id
large_stringlengths
23
28
split
large_stringclasses
1 value
modality
large_stringclasses
1 value
turn_type
large_stringclasses
1 value
dialogue
listlengths
1
8
input
dict
streaming
dict
target
dict
taxonomy
dict
quality
dict
source
dict
metadata
dict
audio_spokenwoz_v02_004440
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "i'm fine. thank you. so.", "turn_id": 2 }, { "role": "user", "text": "i was speaking to albert health food i am traveling to cambridge and i'm looking forward to try the local restaurant i'm l...
{ "chunk_text": "Chunk 1: hello.\nChunk 2: i'm fine. thank you. so.\nChunk 3: i was speaking to albert health food i am traveling to cambridge and i'm looking forward to try the local restaurant i'm looking for a\nChunk 4: restaurant and it should serve british food and it should be in the price in the expensive pric...
{ "num_checkpoints": 5, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "Sure. I can provide the address, phone for the matching restaurant option.", "reasoning": "Across the dialogue, the user is searching for restaurant. The answer should focus on the latest user turn while incorporating relevant context from earlier turns.", "response": "Reasoning: Across the dialogue,...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2025", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_sdeval_v02_000159
train
audio
multi_turn
[ { "role": "user", "text": "How long will it take to get to downtown from here?", "turn_id": 1 } ]
{ "chunk_text": "Chunk 1: How long will it take to get to downtown from here?", "dialogue_text": "Chunk 1: How long will it take to get to downtown from here?", "instruction": "Help the user complete a spoken task-oriented dialogue based on gradually revealed audio/transcript chunks.", "length_bucket": "short",...
{ "num_checkpoints": 1, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "How long will it take to get to downtown from here?", "role": "user", "streaming_reasoning": "The user mentions How long will it take to get to downtown from here?. It i...
{ "answer": "With the current traffic, downtown is approximately 30 minutes away. Would you like the fastest route?", "reasoning": "Across the dialogue, the user discusses \"How long will it take to get to downtown from.\" over 1 turns. The answer should respond to the latest user turn naturally, maintaining the co...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "easy", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SD-Eval", "dialogue_id": "160_3", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004441
train
audio
multi_turn
[ { "role": "user", "text": "oh.", "turn_id": 1 }, { "role": "assistant", "text": "i guess i'm looking for a gesture.", "turn_id": 2 }, { "role": "user", "text": "this went to the city.", "turn_id": 3 }, { "role": "assistant", "text": "yeah.", "turn_id": 4 ...
{ "chunk_text": "Chunk 1: oh.\nChunk 2: i guess i'm looking for a gesture.\nChunk 3: this went to the city.\nChunk 4: yeah.\nChunk 5: but i want to book a table for two people and 12 feet to you on the.\nChunk 6: how about money.", "dialogue_text": "Chunk 1: oh.\nChunk 2: i guess i'm looking for a gesture.\nChunk 3...
{ "num_checkpoints": 6, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "oh.", "role": "user", "streaming_reasoning": "The user mentions Oh. It is not yet clear whether the user needs the full scope of the user's request. The assistant should...
{ "answer": "I can help book that. Could you provide the area, food, price?", "reasoning": "Across the dialogue, the user is searching for gesture. Key context: i guess i'm looking for a gesture; this went to the city; but i want to book a table for two people and 12 feet to you on the; how about money. The answer ...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0635", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004442
train
audio
multi_turn
[ { "role": "user", "text": "mm.", "turn_id": 1 }, { "role": "assistant", "text": "you mentioned that phone number and the present.", "turn_id": 2 }, { "role": "user", "text": "no count. i'm also looking for a train. that she leave one on wednesday.", "turn_id": 3 }, { ...
{ "chunk_text": "Chunk 1: mm.\nChunk 2: you mentioned that phone number and the present.\nChunk 3: no count. i'm also looking for a train. that she leave one on wednesday.\nChunk 4: from cambridge.\nChunk 5: it's not like. it must arrive by 8 8 o'clock.\nChunk 6: in the evening.\nChunk 7: yes.\nChunk 8: so that one w...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "mm.", "role": "user", "streaming_reasoning": "The user mentions Mm. It is not yet clear whether the user needs the full scope of the user's request. The assistant should...
{ "answer": "I can help with the train. Could you share the destination?", "reasoning": "Across the dialogue, the user is searching for train. Key context: you mentioned that phone number and the present; i'm also looking for a train; that she leave one on wednesday; from cambridge. The answer should focus on the l...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2892", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004443
train
audio
multi_turn
[ { "role": "user", "text": "okay, thank you. can you tell me the restaurant number i want to call for some information.", "turn_id": 1 }, { "role": "assistant", "text": "okay, great. thank you. i will call them.", "turn_id": 2 }, { "role": "user", "text": "yes i'm actually als...
{ "chunk_text": "Chunk 1: okay, thank you. can you tell me the restaurant number i want to call for some information.\nChunk 2: okay, great. thank you. i will call them.\nChunk 3: yes i'm actually also looking for a place to go in town uh were you able to find anything like in the same area of the restaurant\nChunk 4...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay, thank you. can you tell me the restaurant number i want to call for some information.", "role": "user", "streaming_reasoning": "The user mentions talking about Oka...
{ "answer": "Sure. I can provide the phone, address for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for to call for some information. Key context: okay, thank you; can you tell me the restaurant number i want to call for some information; okay, great; i will call them....
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "hard", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1868", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004444
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "hi um i want to plan a trip to cambridge and please help me find a train and now live in broxburn and i hope", "turn_id": 2 } ]
{ "chunk_text": "Chunk 1: hello.\nChunk 2: hi um i want to plan a trip to cambridge and please help me find a train and now live in broxburn and i hope", "dialogue_text": "Chunk 1: hello.\nChunk 2: hi um i want to plan a trip to cambridge and please help me find a train and now live in broxburn and i hope", "inst...
{ "num_checkpoints": 2, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "I can help with the train. Could you share the day?", "reasoning": "Across the dialogue, the user discusses \"hello.\" over 2 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across the dialogue, the user discusses \"he...
{ "category": "creative_planning", "subcategory": "event_idea", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0263", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004445
train
audio
multi_turn
[ { "role": "user", "text": "hello. hi, good morning.", "turn_id": 1 }, { "role": "assistant", "text": "hi, good morning. how are you.", "turn_id": 2 }, { "role": "user", "text": "i'm doing great things asking.", "turn_id": 3 }, { "role": "assistant", "text": "u...
{ "chunk_text": "Chunk 1: hello. hi, good morning.\nChunk 2: hi, good morning. how are you.\nChunk 3: i'm doing great things asking.\nChunk 4: uh, could you help me find accommodation average for the weekend.", "dialogue_text": "Chunk 1: hello. hi, good morning.\nChunk 2: hi, good morning. how are you.\nChunk 3: i'...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello. hi, good morning.", "role": "user", "streaming_reasoning": "The user mentions Hello. hi, good morning. It is not yet clear whether the user needs the full scope o...
{ "answer": "I can help. What task or place should I focus on?", "reasoning": "Across the dialogue, the user discusses \"hello. hi, good morning.\" over 4 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across the dialogue, the use...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1233", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001306
train
audio
multi_turn
[ { "role": "user", "text": "Human: Very true! Did you ever play tetris growing up? It was actually designed by a Russian programmer.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: It was one of my favorite games! How about you?", "turn_id": 2 }, { "role": "user", ...
{ "chunk_text": "Chunk 1: Human: Very true! Did you ever play tetris growing up? It was actually designed by a Russian programmer.\nChunk 2: Assistant: It was one of my favorite games! How about you?\nChunk 3: Human: Yes it is so addicting! It is one of the best selling video games of all time! So nice chatting with ...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Very true! Did you ever play tetris growing up? It was actually designed by a Russian programmer.", "role": "user", "streaming_reasoning": "The user mentions talk...
{ "answer": "It has been great! Have a nice night!", "reasoning": "Across the dialogue, the user discusses \"Very true! Did you ever play tetris growing up? It was.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across th...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_bc666142-202c-4473-b378-581a7903ae5f", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001307
train
audio
multi_turn
[ { "role": "user", "text": "Human: Rogue One was kind of refreshing. Good sci-fi with not to crazy of a plot.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Thats was a good one I also liked it, you know Star Wars holds a Guinness World Record for the most successgull film merch...
{ "chunk_text": "Chunk 1: Human: Rogue One was kind of refreshing. Good sci-fi with not to crazy of a plot.\nChunk 2: Assistant: Thats was a good one I also liked it, you know Star Wars holds a Guinness World Record for the most successgull film merchandising film\nChunk 3: Human: I am sure Disney will try hard to br...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Rogue One was kind of refreshing. Good sci-fi with not to crazy of a plot.", "role": "user", "streaming_reasoning": "The user mentions asking for recommendations ...
{ "answer": "Im sure they are getting really involved in Star Wars, they also own the franchise", "reasoning": "Across the dialogue, the user discusses \"Rogue One was kind of refreshing. Good sci-fi with not to.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversati...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_04724774-78d7-4a2c-aebf-7e1116a81c52", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004446
train
audio
multi_turn
[ { "role": "user", "text": "ya um and on top of that i also need to book a taxi to go between the two places and i want to leave", "turn_id": 1 }, { "role": "assistant", "text": "the hotel by uh half past 6 in the morning", "turn_id": 2 }, { "role": "user", "text": "uh, yeah, ...
{ "chunk_text": "Chunk 1: ya um and on top of that i also need to book a taxi to go between the two places and i want to leave\nChunk 2: the hotel by uh half past 6 in the morning\nChunk 3: uh, yeah, so my phone number is 4 5 9.\nChunk 4: 8, 2, 8.\nChunk 5: 0, 3, 2, 3.\nChunk 6: ya. and if you could help me get the c...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "ya um and on top of that i also need to book a taxi to go between the two places and i want to leave", "role": "user", "streaming_reasoning": "The user mentions Ya um an...
{ "answer": "Sure. I can provide the car for the matching taxi option.", "reasoning": "Across the dialogue, the user is wanting to book taxi to go between the two places. Key context: ya um and on top of that i also need to book a taxi to go between the two places and i want to leave the hotel by uh half past 6 in ...
{ "category": "finance_consumer", "subcategory": "budget_planning", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3237", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004447
train
audio
multi_turn
[ { "role": "user", "text": "i also want to look for a place to stay.", "turn_id": 1 }, { "role": "assistant", "text": "uh, yes, i will prefer the the guesthouse.", "turn_id": 2 }, { "role": "user", "text": "and in the moderate price range.", "turn_id": 3 }, { "role...
{ "chunk_text": "Chunk 1: i also want to look for a place to stay.\nChunk 2: uh, yes, i will prefer the the guesthouse.\nChunk 3: and in the moderate price range.\nChunk 4: yes. and uh, it is the best.\nChunk 5: yes. and it should have a sorrow for.\nChunk 6: okay. thank you.\nChunk 7: uh, i want to know whether they...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i also want to look for a place to stay.", "role": "user", "streaming_reasoning": "The user mentions I also want to look for a place to stay. It is not yet clear whether...
{ "answer": "Sure. I can provide the parking, postcode, phone for the matching hotel option.", "reasoning": "Across the dialogue, the user is asking for recommendations about to look for a place to stay. Key context: i also want to look for a place to stay; uh, yes, i will prefer the the guesthouse; and in the mode...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3925", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004448
train
audio
multi_turn
[ { "role": "user", "text": "the ugly ducking case.", "turn_id": 1 }, { "role": "assistant", "text": "yes, please. that's for three people at 8 30 pm.", "turn_id": 2 }, { "role": "user", "text": "on monday.", "turn_id": 3 }, { "role": "assistant", "text": "my na...
{ "chunk_text": "Chunk 1: the ugly ducking case.\nChunk 2: yes, please. that's for three people at 8 30 pm.\nChunk 3: on monday.\nChunk 4: my name is kimberly nichole.\nChunk 5: uh, k imb.\nChunk 6: er l y.\nChunk 7: and my same name is m itc h.\nChunk 8: e double l.", "dialogue_text": "Chunk 1: the ugly ducking ca...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "the ugly ducking case.", "role": "user", "streaming_reasoning": "The user mentions The ugly ducking case. It is not yet clear whether the user needs the full scope of th...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is talking about the ugly ducking case. Key context: the ugly ducking case; yes, please; that's for three people at 8 30 pm; my name is kimberly nichole. The answer should focus on the l...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3587", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001308
train
audio
multi_turn
[ { "role": "user", "text": "Human: Right? Both volleyball and basketball were invented at the YMCA! Let's talk about sharks!", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I learned that, if trained, sharks can bark and that rhymes", "turn_id": 2 }, { "role": "user",...
{ "chunk_text": "Chunk 1: Human: Right? Both volleyball and basketball were invented at the YMCA! Let's talk about sharks!\nChunk 2: Assistant: I learned that, if trained, sharks can bark and that rhymes\nChunk 3: Human: Isn't that something? Have you ever heard of the white shark cafe?", "dialogue_text": "Chunk 1:...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Right? Both volleyball and basketball were invented at the YMCA! Let's talk about sharks!", "role": "user", "streaming_reasoning": "The user mentions talking abou...
{ "answer": "No. Tell me about it", "reasoning": "Across the dialogue, the user discusses \"Right? Both volleyball and basketball were invented at the.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across the dialogue, t...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_52302cea-8208-4161-832f-80a20c0ebf68", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004449
train
audio
multi_turn
[ { "role": "user", "text": "hello, how are you.", "turn_id": 1 }, { "role": "assistant", "text": "i'm more thinking you are speaking to michael and i'm planning a trip to cambridge.", "turn_id": 2 }, { "role": "user", "text": "okay, so basically i'm looking for places to go to...
{ "chunk_text": "Chunk 1: hello, how are you.\nChunk 2: i'm more thinking you are speaking to michael and i'm planning a trip to cambridge.\nChunk 3: okay, so basically i'm looking for places to go to in town. the attraction should be on opposite of the house.\nChunk 4: yes, i want the attraction should it should be ...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello, how are you.", "role": "user", "streaming_reasoning": "The user mentions Hello, how are you. It is not yet clear whether the user needs the full scope of the user...
{ "answer": "I can help find a matching attraction option with those details.", "reasoning": "Across the dialogue, the user is searching for places to go to in town. Key context: hello, how are you; i'm more thinking you are speaking to michael and i'm planning a trip to cambridge; okay, so basically i'm looking fo...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1593", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004450
train
audio
multi_turn
[ { "role": "user", "text": "hi, my name is ellen. i'm planning a trip to cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for a train.", "turn_id": 2 } ]
{ "chunk_text": "Chunk 1: hi, my name is ellen. i'm planning a trip to cambridge.\nChunk 2: i'm looking for a train.", "dialogue_text": "Chunk 1: hi, my name is ellen. i'm planning a trip to cambridge.\nChunk 2: i'm looking for a train.", "instruction": "Help the user complete a spoken task-oriented dialogue base...
{ "num_checkpoints": 2, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, my name is ellen. i'm planning a trip to cambridge.", "role": "user", "streaming_reasoning": "The user mentions Hi, my name is ellen. i'm planning a trip to cambridg...
{ "answer": "I can help with the train. Could you share the departure, destination, day?", "reasoning": "Across the dialogue, the user is searching for train. Key context: hi, my name is ellen; i'm planning a trip to cambridge; i'm looking for a train. The answer should focus on the latest user turn while incorpora...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "easy", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0725", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001309
train
audio
multi_turn
[ { "role": "user", "text": "Human: A 1995 radio station actually pranked Queen Elizabeth. I bet they got in trouble.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Do you still listen to radio? I only listen to streaming service these days.", "turn_id": 2 }, { "role"...
{ "chunk_text": "Chunk 1: Human: A 1995 radio station actually pranked Queen Elizabeth. I bet they got in trouble.\nChunk 2: Assistant: Do you still listen to radio? I only listen to streaming service these days.\nChunk 3: Human: Not really, just in the car. I would listen if Elton John and his song Your Song came on...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: A 1995 radio station actually pranked Queen Elizabeth. I bet they got in trouble.", "role": "user", "streaming_reasoning": "The user mentions talking about 1995, ...
{ "answer": "How yeah? Do you still listen to candle in the wind?", "reasoning": "Across the dialogue, the user discusses \"A 1995 radio station actually pranked Queen Elizabeth. I.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Rea...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_374ae28b-9bc0-45f8-ab39-a52c164c8e79", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004451
train
audio
multi_turn
[ { "role": "user", "text": "got it. thank you so much.", "turn_id": 1 }, { "role": "assistant", "text": "sure. mm. i'm also looking for, uh, trend.", "turn_id": 2 }, { "role": "user", "text": "well, the train should go to cambridge.", "turn_id": 3 }, { "role": "ass...
{ "chunk_text": "Chunk 1: got it. thank you so much.\nChunk 2: sure. mm. i'm also looking for, uh, trend.\nChunk 3: well, the train should go to cambridge.\nChunk 4: and should live on today.\nChunk 5: that's right.\nChunk 6: well, i will depart from london liverpool street.\nChunk 7: and.\nChunk 8: and i and i suppo...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "got it. thank you so much.", "role": "user", "streaming_reasoning": "The user mentions Got it. thank you so much. It is not yet clear whether the user needs the full sco...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is searching for thank you so much. Key context: thank you so much; i'm also looking for, uh, trend; well, the train should go to cambridge; and should live on today. The answer should focus on the...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0338", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004452
train
audio
multi_turn
[ { "role": "user", "text": "bread and their address.", "turn_id": 1 }, { "role": "assistant", "text": "kelly and how will i find.", "turn_id": 2 }, { "role": "user", "text": "thank you. i have the number, please.", "turn_id": 3 }, { "role": "assistant", "text":...
{ "chunk_text": "Chunk 1: bread and their address.\nChunk 2: kelly and how will i find.\nChunk 3: thank you. i have the number, please.\nChunk 4: i will confirm it. is it 1, 2, 2.\nChunk 5: 3, 3, 5.\nChunk 6: 8 9 6 6.\nChunk 7: thank you. you've been great.\nChunk 8: that would be all.", "dialogue_text": "Chunk 1: ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "bread and their address.", "role": "user", "streaming_reasoning": "The user mentions Bread and their address. It is not yet clear whether the user needs the full scope o...
{ "answer": "Sure. I can provide the address, phone for the matching police option.", "reasoning": "Across the dialogue, the user is searching for bread and their address. Key context: bread and their address; kelly and how will i find; i have the number, please; i will confirm it. The answer should focus on the la...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1139", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004453
train
audio
multi_turn
[ { "role": "user", "text": "um that's okay. thanks.", "turn_id": 1 }, { "role": "assistant", "text": "he has been for five people. on monday at 1130 in the morning.", "turn_id": 2 }, { "role": "user", "text": "my name is william hammond.", "turn_id": 3 }, { "role":...
{ "chunk_text": "Chunk 1: um that's okay. thanks.\nChunk 2: he has been for five people. on monday at 1130 in the morning.\nChunk 3: my name is william hammond.\nChunk 4: ha double m.\nChunk 5: o nd.\nChunk 6: yes, please. i'm also looking for a train.\nChunk 7: the train should be. should depart from knowledge going...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "um that's okay. thanks.", "role": "user", "streaming_reasoning": "The user mentions Um that's okay. thanks. It is not yet clear whether the user needs the full scope of ...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is searching for train. Key context: um that's okay; he has been for five people; on monday at 1130 in the morning; my name is william hammond. The answer should focus on the latest user turn while...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2043", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004454
train
audio
multi_turn
[ { "role": "user", "text": "hi, my name is ellen. i'm planning a trip to cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for a train.", "turn_id": 2 }, { "role": "user", "text": "from birmingham new street to cambridge.", "turn_id": 3 }, { ...
{ "chunk_text": "Chunk 1: hi, my name is ellen. i'm planning a trip to cambridge.\nChunk 2: i'm looking for a train.\nChunk 3: from birmingham new street to cambridge.\nChunk 4: sure thing. thank you very much.", "dialogue_text": "Chunk 1: hi, my name is ellen. i'm planning a trip to cambridge.\nChunk 2: i'm lookin...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, my name is ellen. i'm planning a trip to cambridge.", "role": "user", "streaming_reasoning": "The user mentions Hi, my name is ellen. i'm planning a trip to cambridg...
{ "answer": "I can help with the train. Could you share the day, time?", "reasoning": "Across the dialogue, the user is searching for train. Key context: hi, my name is ellen; i'm planning a trip to cambridge; i'm looking for a train; from birmingham new street to cambridge. The answer should focus on the latest us...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0725", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004455
train
audio
multi_turn
[ { "role": "user", "text": "then i will settle for the moderate price range.", "turn_id": 1 }, { "role": "assistant", "text": "mm. okay, it will be perfect.", "turn_id": 2 }, { "role": "user", "text": "yes, book it for two people. uh, and five nights.", "turn_id": 3 }, ...
{ "chunk_text": "Chunk 1: then i will settle for the moderate price range.\nChunk 2: mm. okay, it will be perfect.\nChunk 3: yes, book it for two people. uh, and five nights.\nChunk 4: i. which is best of the friday.\nChunk 5: yes, perfect.\nChunk 6: if i wish you get h for hannah.\nChunk 7: e for elephant double l f...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "then i will settle for the moderate price range.", "role": "user", "streaming_reasoning": "The user mentions Then i will settle for the moderate price range. It is not y...
{ "answer": "I can help book that. Could you provide the area, hotel_type?", "reasoning": "Across the dialogue, the user is wanting to book it for two people. Key context: then i will settle for the moderate price range; okay, it will be perfect; yes, book it for two people; uh, and five nights. The answer should f...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2888", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004456
train
audio
multi_turn
[ { "role": "user", "text": "oh. hello.", "turn_id": 1 }, { "role": "assistant", "text": "hi, i'm looking for four star hotel in the expensive price range that also has free wifi.", "turn_id": 2 }, { "role": "user", "text": "okay. can i book the han jing jian marriott hotel for...
{ "chunk_text": "Chunk 1: oh. hello.\nChunk 2: hi, i'm looking for four star hotel in the expensive price range that also has free wifi.\nChunk 3: okay. can i book the han jing jian marriott hotel for three people three nights starting wednesday.\nChunk 4: oh, okay. how about monday.\nChunk 5: sure. my name is gino, ...
{ "num_checkpoints": 7, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "oh. hello.", "role": "user", "streaming_reasoning": "The user mentions Oh. hello. It is not yet clear whether the user needs the full scope of the user's request. The as...
{ "answer": "I can help book that. Could you provide the area, hotel_type?", "reasoning": "Across the dialogue, the user is searching for four star hotel in the expensive price range that also has free wifi. Key context: hi, i'm looking for four star hotel in the expensive price range that also has free wifi; can i...
{ "category": "tech_support", "subcategory": "device_troubleshooting", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3238", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004457
train
audio
multi_turn
[ { "role": "user", "text": "hi, hello.", "turn_id": 1 }, { "role": "assistant", "text": "good in yourself.", "turn_id": 2 }, { "role": "user", "text": "um i need your assistance. i'm looking for a restaurant.", "turn_id": 3 }, { "role": "assistant", "text": "so...
{ "chunk_text": "Chunk 1: hi, hello.\nChunk 2: good in yourself.\nChunk 3: um i need your assistance. i'm looking for a restaurant.\nChunk 4: so the restaurant should serve indian food and should be in the west and it did have an expensive price range.\nChunk 5: okay. um, can i please get the phone number.\nChunk 6: ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, hello.", "role": "user", "streaming_reasoning": "The user mentions Hi, hello. It is not yet clear whether the user needs the full scope of the user's request. The as...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is searching for restaurant. Key context: good in yourself; um i need your assistance; i'm looking for a restaurant; so the restaurant should serve indian food and should be in the west ...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2222", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001310
train
audio
multi_turn
[ { "role": "user", "text": "Human: Yeah I agree, so do you also like soccer? its a sport that moves a lot of money", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I have watched soccer on occassion, but I do not keep up with it. Do like soccer?", "turn_id": 2 }, { "ro...
{ "chunk_text": "Chunk 1: Human: Yeah I agree, so do you also like soccer? its a sport that moves a lot of money\nChunk 2: Assistant: I have watched soccer on occassion, but I do not keep up with it. Do like soccer?\nChunk 3: Human: I do like soccer, there is a German player that donated his winning bonus to 23 Brazi...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Yeah I agree, so do you also like soccer? its a sport that moves a lot of money", "role": "user", "streaming_reasoning": "The user mentions talking about Human, Y...
{ "answer": "Wow that is amazing!", "reasoning": "Across the dialogue, the user discusses \"Yeah I agree, so do you also like soccer? its a sport that.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across the dialogue, t...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_81ea9ea5-f9ee-4b15-8764-940268e345c1", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004458
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "yes, i am traveling to cambridge at the moment. i need to find some information.", "turn_id": 2 }, { "role": "user", "text": "so i am looking for a restaurant. uh, i have the name would you ab...
{ "chunk_text": "Chunk 1: hello.\nChunk 2: yes, i am traveling to cambridge at the moment. i need to find some information.\nChunk 3: so i am looking for a restaurant. uh, i have the name would you able to check if i tell the name.\nChunk 4: so the name is called dairy house and wine shop.\nChunk 5: yes, there are. t...
{ "num_checkpoints": 5, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is searching for restaurant. Key context: yes, i am traveling to cambridge at the moment; i need to find some information; so i am looking for a restaurant; uh, i have the name would you...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1887", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004459
train
audio
multi_turn
[ { "role": "user", "text": "my id number is 0 9 2 0.", "turn_id": 1 }, { "role": "assistant", "text": "8 0 8.", "turn_id": 2 }, { "role": "user", "text": "at 6 7.", "turn_id": 3 }, { "role": "assistant", "text": "what in the night 7 5.", "turn_id": 4 }, ...
{ "chunk_text": "Chunk 1: my id number is 0 9 2 0.\nChunk 2: 8 0 8.\nChunk 3: at 6 7.\nChunk 4: what in the night 7 5.\nChunk 5: i'm also looking for a place to stay.\nChunk 6: i'm looking for a hotel that in the center.\nChunk 7: and it should include free parking spot and free wi fi.\nChunk 8: okay, i would like to...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "my id number is 0 9 2 0.", "role": "user", "streaming_reasoning": "The user mentions My id number is 0 9 2 0. It is not yet clear whether the user needs the full scope o...
{ "answer": "I can help book that. Could you provide the price, hotel_type, day?", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: my id number is 0 9 2 0; what in the night 7 5; i'm also looking for a place to stay; i'm looking for a hotel that in the center. The answer sho...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2653", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004460
train
audio
multi_turn
[ { "role": "user", "text": "good day. i am looking for information about a train in cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i am looking for a train that departs from cambridge and goes to stanford airport.", "turn_id": 2 }, { "role": "user", "text": "and ...
{ "chunk_text": "Chunk 1: good day. i am looking for information about a train in cambridge.\nChunk 2: i am looking for a train that departs from cambridge and goes to stanford airport.\nChunk 3: and it should leave after. a quota. 8 pm on monday.\nChunk 4: okay, i would like to book 5.5 egg.\nChunk 5: my id number i...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "good day. i am looking for information about a train in cambridge.", "role": "user", "streaming_reasoning": "The user mentions searching for Good. It is not yet clear wh...
{ "answer": "I can help book that with the details you gave. Would you like me to confirm the reservation?", "reasoning": "Across the dialogue, the user is searching for information about a train in cambridge. Key context: i am looking for information about a train in cambridge; i am looking for a train that depart...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2653", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004461
train
audio
multi_turn
[ { "role": "user", "text": "and the mead.", "turn_id": 1 }, { "role": "assistant", "text": "in the midst of town.", "turn_id": 2 }, { "role": "user", "text": "thing is.", "turn_id": 3 }, { "role": "assistant", "text": "free packing.", "turn_id": 4 }, { ...
{ "chunk_text": "Chunk 1: and the mead.\nChunk 2: in the midst of town.\nChunk 3: thing is.\nChunk 4: free packing.\nChunk 5: okay.\nChunk 6: yes, i'm still here.\nChunk 7: mm.\nChunk 8: it's not perfect. i think i'm going for that one.", "dialogue_text": "Chunk 1: and the mead.\nChunk 2: in the midst of town.\nChu...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "and the mead.", "role": "user", "streaming_reasoning": "The user mentions And the mead. It is not yet clear whether the user needs the full scope of the user's request. ...
{ "answer": "I can help with the hotel. Could you share the price, hotel_type?", "reasoning": "Across the dialogue, the user is talking about and the mead. Key context: and the mead; in the midst of town; free packing; yes, i'm still here. The answer should focus on the latest user turn while incorporating relevant...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "opinion_sharing" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3516", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001311
train
audio
multi_turn
[ { "role": "user", "text": "Human: LOL In the 1960's top bowlers made twice as much as top football stars of the day. I'm sure that's way different now.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Bowlers are making a comeback. They are seeing more and more concussions these ...
{ "chunk_text": "Chunk 1: Human: LOL In the 1960's top bowlers made twice as much as top football stars of the day. I'm sure that's way different now.\nChunk 2: Assistant: Bowlers are making a comeback. They are seeing more and more concussions these days! Soon they will be putting tracking chips in their bowling bal...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: LOL In the 1960's top bowlers made twice as much as top football stars of the day. I'm sure that's way different now.", "role": "user", "streaming_reasoning": "Th...
{ "answer": "222-0... That must have been a brutal game!", "reasoning": "Across the dialogue, the user is asking for help with human: lol in the 1960's top bowlers made twice as much as top football stars of the day. Key context: Human: LOL In the 1960's top bowlers made twice as much as top football stars of the d...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_c6035b2e-a8f9-4c1c-9f2f-f4e5a6035c99", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004462
train
audio
multi_turn
[ { "role": "user", "text": "hello. hello.", "turn_id": 1 }, { "role": "assistant", "text": "ya um so i'm looking for a place to stay uh it should include free parking and it should be a two star uh hotel", "turn_id": 2 }, { "role": "user", "text": "in the north", "turn_id"...
{ "chunk_text": "Chunk 1: hello. hello.\nChunk 2: ya um so i'm looking for a place to stay uh it should include free parking and it should be a two star uh hotel\nChunk 3: in the north\nChunk 4: okay. um which one is cheaper for a double room.", "dialogue_text": "Chunk 1: hello. hello.\nChunk 2: ya um so i'm lookin...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello. hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. hello. It is not yet clear whether the user needs the full scope of the user's request. ...
{ "answer": "I can help with the hotel. Could you share the price?", "reasoning": "Across the dialogue, the user is searching for place to stay uh it should include free parking. Key context: ya um so i'm looking for a place to stay uh it should include free parking and it should be a two star uh hotel in the north...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3225", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004463
train
audio
multi_turn
[ { "role": "user", "text": "uh, i will go with the a. u.d. c theater.", "turn_id": 1 }, { "role": "assistant", "text": "um i want to know the address of this theater.", "turn_id": 2 }, { "role": "user", "text": "okay. and how about the phone number of this place.", "turn_i...
{ "chunk_text": "Chunk 1: uh, i will go with the a. u.d. c theater.\nChunk 2: um i want to know the address of this theater.\nChunk 3: okay. and how about the phone number of this place.\nChunk 4: okay, i'll call them to understand more first. thank you.\nChunk 5: uh, can you also help me to buy a taxi to travel betw...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "uh, i will go with the a. u.d. c theater.", "role": "user", "streaming_reasoning": "The user mentions Uh, i will go with the a. u.d. c theater. It is not yet clear wheth...
{ "answer": "I can help with the taxi. Could you share the departure, destination?", "reasoning": "Across the dialogue, the user is asking for help with to know the address of this theater. Key context: uh, i will go with the a; um i want to know the address of this theater; and how about the phone number of this p...
{ "category": "finance_consumer", "subcategory": "budget_planning", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1903", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004464
train
audio
multi_turn
[ { "role": "user", "text": "hi, i'm looking for a, uh. ta ke yi.", "turn_id": 1 }, { "role": "assistant", "text": "uh, i'm departing from kingsland train station.", "turn_id": 2 }, { "role": "user", "text": "uh, i will go to the partial restaurant and i hope it could arrive by...
{ "chunk_text": "Chunk 1: hi, i'm looking for a, uh. ta ke yi.\nChunk 2: uh, i'm departing from kingsland train station.\nChunk 3: uh, i will go to the partial restaurant and i hope it could arrive by 1230 in the morning.\nChunk 4: and my phone number is 3 6 1.\nChunk 5: 0 1 9.\nChunk 6: 5, 3, 2, 3.", "dialogue_tex...
{ "num_checkpoints": 6, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, i'm looking for a, uh. ta ke yi.", "role": "user", "streaming_reasoning": "The user mentions Hi, i'm looking for a, uh. ta ke yi. It is not yet clear whether the use...
{ "answer": "I can help find a matching taxi option with those details.", "reasoning": "Across the dialogue, the user is searching for hi, i'm looking for a, uh. Key context: hi, i'm looking for a, uh; uh, i'm departing from kingsland train station; uh, i will go to the partial restaurant and i hope it could arrive...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0103", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004465
train
audio
multi_turn
[ { "role": "user", "text": "9th and one more thing is just.", "turn_id": 1 }, { "role": "assistant", "text": "i prefer taxi.", "turn_id": 2 }, { "role": "user", "text": "it should arrive at the restaurant by 1015.", "turn_id": 3 }, { "role": "assistant", "text"...
{ "chunk_text": "Chunk 1: 9th and one more thing is just.\nChunk 2: i prefer taxi.\nChunk 3: it should arrive at the restaurant by 1015.\nChunk 4: my number is 1 1 4.\nChunk 5: 3, 3, 4.\nChunk 6: 1 9 2 6.\nChunk 7: correct. and what is the car type.\nChunk 8: thank you very much for your help.", "dialogue_text": "C...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "9th and one more thing is just.", "role": "user", "streaming_reasoning": "The user mentions 9th and one more thing is just. It is not yet clear whether the user needs th...
{ "answer": "Sure. I can provide the car for the matching taxi option.", "reasoning": "Across the dialogue, the user is asking about 9th and one more thing is just. Key context: 9th and one more thing is just; i prefer taxi; it should arrive at the restaurant by 1015; my number is 1 1 4. The answer should focus on ...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3796", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001312
train
audio
multi_turn
[ { "role": "user", "text": "Human: And on the opposite end of the spectrum, we have Baker Mayfield. From quarterback to professional gamer... sign me up!", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Yes absolutely! He definitely made the right choice going into the NFL, do you...
{ "chunk_text": "Chunk 1: Human: And on the opposite end of the spectrum, we have Baker Mayfield. From quarterback to professional gamer... sign me up!\nChunk 2: Assistant: Yes absolutely! He definitely made the right choice going into the NFL, do you know how ESPN won an Emmy though?\nChunk 3: Human: No! That's awes...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: And on the opposite end of the spectrum, we have Baker Mayfield. From quarterback to professional gamer... sign me up!", "role": "user", "streaming_reasoning": "T...
{ "answer": "Yes I like bowling two, I have a better chance of being a professional bowler than being an NFL star haha.", "reasoning": "Across the dialogue, the user discusses \"And on the opposite end of the spectrum, we have Baker.\" over 3 turns. The answer should respond to the latest user turn naturally, maint...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_42493ab0-cd52-4e22-a9de-93359b195e06", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004466
train
audio
multi_turn
[ { "role": "user", "text": "yes, of course, my name is chelsea mcconnell.", "turn_id": 1 }, { "role": "assistant", "text": "it's ç heelseachelsea. mahoneymahoney.", "turn_id": 2 }, { "role": "user", "text": "yes. uh, well, i'm also looking for a particular attraction. can you ...
{ "chunk_text": "Chunk 1: yes, of course, my name is chelsea mcconnell.\nChunk 2: it's ç heelseachelsea. mahoneymahoney.\nChunk 3: yes. uh, well, i'm also looking for a particular attraction. can you help me.\nChunk 4: i remember it called the face club.\nChunk 5: yes. uh, before that, can you tell me some informatio...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes, of course, my name is chelsea mcconnell.", "role": "user", "streaming_reasoning": "The user mentions Yes, of course, my name is chelsea mcconnell. It is not yet cle...
{ "answer": "Sure. I can provide the fee, postcode for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for particular attraction. Key context: yes, of course, my name is chelsea mcconnell; it's ç heelseachelsea; mahoneymahoney; uh, well, i'm also looking for a particular a...
{ "category": "education", "subcategory": "course_planning", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2295", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004467
train
audio
multi_turn
[ { "role": "user", "text": "i am good. i am looking for information in cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for a place to stay.", "turn_id": 2 }, { "role": "user", "text": "i am looking for a hotel.", "turn_id": 3 }, { "role": "...
{ "chunk_text": "Chunk 1: i am good. i am looking for information in cambridge.\nChunk 2: i'm looking for a place to stay.\nChunk 3: i am looking for a hotel.\nChunk 4: i want it to be in a moderate. price range.\nChunk 5: a star of four.\nChunk 6: for free parking and that is off.\nChunk 7: free parking, that's all....
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i am good. i am looking for information in cambridge.", "role": "user", "streaming_reasoning": "The user mentions I am good. i am looking for information in cambridge. I...
{ "answer": "I can help with the hotel. Could you share the area?", "reasoning": "Across the dialogue, the user is searching for information in cambridge. Key context: i am looking for information in cambridge; i'm looking for a place to stay; i am looking for a hotel; i want it to be in a moderate. The answer shou...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3094", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004468
train
audio
multi_turn
[ { "role": "user", "text": "okay. and what's the phone number.", "turn_id": 1 }, { "role": "assistant", "text": "okay, thank you. and i'm also looking for a particular hotel.", "turn_id": 2 }, { "role": "user", "text": "yeah, this work works.", "turn_id": 3 }, { "r...
{ "chunk_text": "Chunk 1: okay. and what's the phone number.\nChunk 2: okay, thank you. and i'm also looking for a particular hotel.\nChunk 3: yeah, this work works.\nChunk 4: okay, thank you. and i want to book a room for two people.\nChunk 5: uh, i like to stay from sunday to wednesday.\nChunk 6: yes. uh, sure, my ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay. and what's the phone number.", "role": "user", "streaming_reasoning": "The user mentions Okay. and what's the phone number. It is not yet clear whether the user ne...
{ "answer": "I can help book that. Could you provide the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for particular hotel. Key context: and what's the phone number; okay, thank you; and i'm also looking for a particular hotel; yeah, this work works. The answer should focus on...
{ "category": "work_productivity", "subcategory": "meeting_summary", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0684", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004469
train
audio
multi_turn
[ { "role": "user", "text": "no. er.", "turn_id": 1 }, { "role": "assistant", "text": "p n.", "turn_id": 2 }, { "role": "user", "text": "and i am at at outlook dot com.", "turn_id": 3 }, { "role": "assistant", "text": "yes, i'm also looking for a trainer.", ...
{ "chunk_text": "Chunk 1: no. er.\nChunk 2: p n.\nChunk 3: and i am at at outlook dot com.\nChunk 4: yes, i'm also looking for a trainer.\nChunk 5: the train should be on the same day as order booking. and should depart from stephen.\nChunk 6: yes, ma'am.\nChunk 7: yes, the train should go to cambridge and should lea...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "no. er.", "role": "user", "streaming_reasoning": "The user mentions No. er. It is not yet clear whether the user needs the full scope of the user's request. The assistan...
{ "answer": "I can help book that. Could you provide the time?", "reasoning": "Across the dialogue, the user is searching for trainer. Key context: and i am at at outlook dot com; yes, i'm also looking for a trainer; the train should be on the same day as order booking; and should depart from stephen. The answer sh...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1101", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_sdeval_v02_000160
train
audio
multi_turn
[ { "role": "user", "text": "Can you suggest a good podcast?", "turn_id": 1 } ]
{ "chunk_text": "Chunk 1: Can you suggest a good podcast?", "dialogue_text": "Chunk 1: Can you suggest a good podcast?", "instruction": "Help the user complete a spoken task-oriented dialogue based on gradually revealed audio/transcript chunks.", "length_bucket": "short", "num_turns": 1, "total_transcript_t...
{ "num_checkpoints": 1, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Can you suggest a good podcast?", "role": "user", "streaming_reasoning": "The user mentions Can you suggest a good podcast?. It is not yet clear whether the user needs w...
{ "answer": "Absolutely, how about 'This American Life'? It offers interesting stories and diverse topics to keep you engaged.", "reasoning": "Across the dialogue, the user discusses \"Can you suggest a good.\" over 1 turns. The answer should respond to the latest user turn naturally, maintaining the conversational...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "easy", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SD-Eval", "dialogue_id": "56_0", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001313
train
audio
multi_turn
[ { "role": "user", "text": "Human: Yeah, I hear Michael Giacchino was back to score again.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Yup, and I just love the work of Holly Hunter and Craig T Nelson too!", "turn_id": 2 }, { "role": "user", "text": "Human: The...
{ "chunk_text": "Chunk 1: Human: Yeah, I hear Michael Giacchino was back to score again.\nChunk 2: Assistant: Yup, and I just love the work of Holly Hunter and Craig T Nelson too!\nChunk 3: Human: They are pretty incredible. No pun intended.", "dialogue_text": "Chunk 1: Human: Yeah, I hear Michael Giacchino was bac...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Yeah, I hear Michael Giacchino was back to score again.", "role": "user", "streaming_reasoning": "The user mentions talking about Human, Yeah, Michael Giacchino. ...
{ "answer": "Did you know that there was a comic series written about cleaning up collateral damage caused by superheroes? Pretty creative!", "reasoning": "Across the dialogue, the user discusses \"Yeah, I hear Michael Giacchino was back to score.\" over 3 turns. The answer should respond to the latest user turn na...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_76ceb346-3073-45af-bc41-f7152dc35fc8", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001314
train
audio
multi_turn
[ { "role": "user", "text": "Human: Hey! Did you see Wonder Woman when it came out (or since then)?", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I have not seen the new WW movie. Have you?", "turn_id": 2 }, { "role": "user", "text": "Human: I did see it. I'm not...
{ "chunk_text": "Chunk 1: Human: Hey! Did you see Wonder Woman when it came out (or since then)?\nChunk 2: Assistant: I have not seen the new WW movie. Have you?\nChunk 3: Human: I did see it. I'm not a big fan of superhero movies in general but I wanted to support movies with female superhero leads", "dialogue_tex...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Hey! Did you see Wonder Woman when it came out (or since then)?", "role": "user", "streaming_reasoning": "The user mentions talking about Human, Wonder Woman. It ...
{ "answer": "Yeah, that is a good idea. i figure that women heroes need to be more visible. i usually dont watch those movies because the explosions and battle scenes get dull.", "reasoning": "Across the dialogue, the user discusses \"Hey! Did you see Wonder Woman when it came out (or since.\" over 3 turns. The ans...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_cd5b3f34-1274-4dcd-92b8-9b804615e5ad", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001315
train
audio
multi_turn
[ { "role": "user", "text": "Human: Dang I wonder how many trips he took moving all of those!", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: That is a good point, 32,000 books sounds like a heck of a lot of wheelbarrow runs!", "turn_id": 2 }, { "role": "user", "te...
{ "chunk_text": "Chunk 1: Human: Dang I wonder how many trips he took moving all of those!\nChunk 2: Assistant: That is a good point, 32,000 books sounds like a heck of a lot of wheelbarrow runs!\nChunk 3: Human: It does! If I release a book, I'm going to do it in Norway because you're bound to sell at least 1000 cop...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Dang I wonder how many trips he took moving all of those!", "role": "user", "streaming_reasoning": "The user mentions talking about Human, Dang. It is not yet cle...
{ "answer": "That is pretty wild, the government stepping in and buying them to distribute to libraries around the country, cool.", "reasoning": "Across the dialogue, the user is wanting to book human: dang i wonder how many trips he took moving all of those. Key context: Human: Dang I wonder how many trips he took...
{ "category": "information_extraction", "subcategory": "long_note_to_summary", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_3fe97ee6-ab6e-4d1c-9306-2c76d10740ee", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004470
train
audio
multi_turn
[ { "role": "user", "text": "yes i'm looking for a place to dine perhaps something in spanish food in the center of town if possible and um you know it", "turn_id": 1 }, { "role": "assistant", "text": "could be an expensive range", "turn_id": 2 }, { "role": "user", "text": "wel...
{ "chunk_text": "Chunk 1: yes i'm looking for a place to dine perhaps something in spanish food in the center of town if possible and um you know it\nChunk 2: could be an expensive range\nChunk 3: well, how about indian food.\nChunk 4: oh, great. could i get the post code for that.\nChunk 5: oh great thank you um i'm...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes i'm looking for a place to dine perhaps something in spanish food in the center of town if possible and um you know it", "role": "user", "streaming_reasoning": "The ...
{ "answer": "Sure. I can provide the fee for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for yes i'm looking for a place to dine perhaps something in spanish food in the center of town if possible and um you know it could be an expensive range well, how about indian fo...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "hard", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3952", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004471
train
audio
multi_turn
[ { "role": "user", "text": "yes, it should be in the west. and also serve some indian food.", "turn_id": 1 }, { "role": "assistant", "text": "okay. um, may you please, um. book a table for me and seven other people, please.", "turn_id": 2 }, { "role": "user", "text": "uh could...
{ "chunk_text": "Chunk 1: yes, it should be in the west. and also serve some indian food.\nChunk 2: okay. um, may you please, um. book a table for me and seven other people, please.\nChunk 3: uh could you please make it for saturday and it should be um at um it should be 30 minutes uh before 3 3 uh 3 pm\nChunk 4: oka...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes, it should be in the west. and also serve some indian food.", "role": "user", "streaming_reasoning": "The user mentions Yes, it should be in the west. It is not yet ...
{ "answer": "I can help with the hotel. Could you share the price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: yes, it should be in the west; and also serve some indian food; um, may you please, um; book a table for me and seven other people, please. The an...
{ "category": "tech_support", "subcategory": "device_troubleshooting", "difficulty": "hard", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2471", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004472
train
audio
multi_turn
[ { "role": "user", "text": "morning. my name is joella. i'm looking for a place to dine in cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for an expensive place that serves. let's say, indian food in cambridge.", "turn_id": 2 } ]
{ "chunk_text": "Chunk 1: morning. my name is joella. i'm looking for a place to dine in cambridge.\nChunk 2: i'm looking for an expensive place that serves. let's say, indian food in cambridge.", "dialogue_text": "Chunk 1: morning. my name is joella. i'm looking for a place to dine in cambridge.\nChunk 2: i'm look...
{ "num_checkpoints": 2, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "morning. my name is joella. i'm looking for a place to dine in cambridge.", "role": "user", "streaming_reasoning": "The user mentions searching for Morning. It is not ye...
{ "answer": "I can help with the restaurant. Could you share the area?", "reasoning": "Across the dialogue, the user is searching for place to dine in cambridge. Key context: my name is joella; i'm looking for a place to dine in cambridge; i'm looking for an expensive place that serves; let's say, indian food in ca...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1569", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004473
train
audio
multi_turn
[ { "role": "user", "text": "okay thank you and i'm also looking for a train so i want to go to cambridge um departing from the station with initial lkc", "turn_id": 1 }, { "role": "assistant", "text": "i want to leave on wednesday and arrive by 1 30 pm", "turn_id": 2 }, { "role": ...
{ "chunk_text": "Chunk 1: okay thank you and i'm also looking for a train so i want to go to cambridge um departing from the station with initial lkc\nChunk 2: i want to leave on wednesday and arrive by 1 30 pm\nChunk 3: ya, that's right.\nChunk 4: okay. um, can i just find out about the latest strain.\nChunk 5: okay...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay thank you and i'm also looking for a train so i want to go to cambridge um departing from the station with initial lkc", "role": "user", "streaming_reasoning": "The...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is searching for okay thank you and i'm also looking for a train so i want to go to cambridge um departing from the station with initial lkc i want to leave on wednesday and arrive by 1 30 pm ya, t...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3257", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004474
train
audio
multi_turn
[ { "role": "user", "text": "hello. is this customer service.", "turn_id": 1 }, { "role": "assistant", "text": "i'm currently traveling in cambridge and i have some problems.", "turn_id": 2 }, { "role": "user", "text": "i'm trying to find a restaurant to dine.", "turn_id": ...
{ "chunk_text": "Chunk 1: hello. is this customer service.\nChunk 2: i'm currently traveling in cambridge and i have some problems.\nChunk 3: i'm trying to find a restaurant to dine.\nChunk 4: uh, the location of the hotel has to be in the west of the city.", "dialogue_text": "Chunk 1: hello. is this customer servi...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello. is this customer service.", "role": "user", "streaming_reasoning": "The user mentions Hello. is this customer service. It is not yet clear whether the user needs ...
{ "answer": "I can help with the restaurant. Could you share the food, price?", "reasoning": "Across the dialogue, the user discusses \"hello. is this customer.\" over 4 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.", "response": "Reasoning: Across the di...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0232", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001316
train
audio
multi_turn
[ { "role": "user", "text": "Human: Oh, I've never heard of them but they are geniuses for sure! Have you ever read the series about fixing the collateral damage caused by the superheroes?", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I have not read that series. I have read a l...
{ "chunk_text": "Chunk 1: Human: Oh, I've never heard of them but they are geniuses for sure! Have you ever read the series about fixing the collateral damage caused by the superheroes?\nChunk 2: Assistant: I have not read that series. I have read a lot of comics from Earth -616 which was Marvels original comic book ...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Oh, I've never heard of them but they are geniuses for sure! Have you ever read the series about fixing the collateral damage caused by the superheroes?", "role": "user...
{ "answer": "Garfield was in Pet Force, a comic that didn't go very far. They say there are over 300 registered superheroes in the U.S.but I think that sounds kind of low.", "reasoning": "Across the dialogue, the user is wanting to book universe. Key context: Human: Oh, I've never heard of them but they are geniuse...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_9345649c-9bb1-470a-a022-9ee33e1c7eba", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004475
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "yes, i'm traveling to cambridge at the moment and i want i need help in finding some information.", "turn_id": 2 }, { "role": "user", "text": "i'm looking for a place to eat. do you have anyth...
{ "chunk_text": "Chunk 1: hello.\nChunk 2: yes, i'm traveling to cambridge at the moment and i want i need help in finding some information.\nChunk 3: i'm looking for a place to eat. do you have anything, uh, in an expensive price price range to recommend.\nChunk 4: uh, please try to find anything west side of the ci...
{ "num_checkpoints": 6, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "I can help with the restaurant. Could you share the price?", "reasoning": "Across the dialogue, the user is searching for place to eat. Key context: yes, i'm traveling to cambridge at the moment and i want i need help in finding some information; i'm looking for a place to eat; do you have anything, uh...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1868", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004476
train
audio
multi_turn
[ { "role": "user", "text": "hello. is that customer service.", "turn_id": 1 }, { "role": "assistant", "text": "uh. i'm looking for a restaurant.", "turn_id": 2 } ]
{ "chunk_text": "Chunk 1: hello. is that customer service.\nChunk 2: uh. i'm looking for a restaurant.", "dialogue_text": "Chunk 1: hello. is that customer service.\nChunk 2: uh. i'm looking for a restaurant.", "instruction": "Help the user complete a spoken task-oriented dialogue based on gradually revealed audi...
{ "num_checkpoints": 2, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello. is that customer service.", "role": "user", "streaming_reasoning": "The user mentions Hello. is that customer service. It is not yet clear whether the user needs ...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is searching for restaurant. Key context: is that customer service; i'm looking for a restaurant. The answer should focus on the latest user turn while incorporating relevant context fro...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "easy", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0277", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004477
train
audio
multi_turn
[ { "role": "user", "text": "i'm looking for a train.", "turn_id": 1 }, { "role": "assistant", "text": "i will depart from birmingham new street.", "turn_id": 2 }, { "role": "user", "text": "i'm headed for cambridge.", "turn_id": 3 }, { "role": "assistant", "tex...
{ "chunk_text": "Chunk 1: i'm looking for a train.\nChunk 2: i will depart from birmingham new street.\nChunk 3: i'm headed for cambridge.\nChunk 4: i would love to leave on sunday after quarter past 4.\nChunk 5: i will hold thank you.\nChunk 6: you are welcome.\nChunk 7: to rate and can i have some information.\nChu...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i'm looking for a train.", "role": "user", "streaming_reasoning": "The user mentions I'm looking for a train. It is not yet clear whether the user needs the full scope o...
{ "answer": "Sure. I can provide the arriveby for the matching train option.", "reasoning": "Across the dialogue, the user is searching for train. Key context: i'm looking for a train; i will depart from birmingham new street; i'm headed for cambridge; i would love to leave on sunday after quarter past 4. The answe...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1151", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001317
train
audio
multi_turn
[ { "role": "user", "text": "Human: Well, I guess the only way to find out is to read the book! LOL What other fiction do you like?", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I read all of the Black Stallion fiction books as a kid. I had a lot of books I liked but really can'...
{ "chunk_text": "Chunk 1: Human: Well, I guess the only way to find out is to read the book! LOL What other fiction do you like?\nChunk 2: Assistant: I read all of the Black Stallion fiction books as a kid. I had a lot of books I liked but really can't remember them all! Which fiction books do you like?\nChunk 3: Hum...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Well, I guess the only way to find out is to read the book! LOL What other fiction do you like?", "role": "user", "streaming_reasoning": "The user mentions search...
{ "answer": "I still haven't read the LOTR books, unfortunately. I was waiting to get some type of special collector edition from a local bookstore that sells lots of classic fiction books. I love the movies, too.", "reasoning": "Across the dialogue, the user is searching for human: well, i guess the only way to fi...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "hard", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_cd486bd3-6d8f-4c10-b83c-1a89d351e2cd", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004478
train
audio
multi_turn
[ { "role": "user", "text": "hello, my name is emily. i'm looking for a place to stay.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for, uh, for one which is, uh, in the expensive price inch.", "turn_id": 2 }, { "role": "user", "text": "okay, i'm looking at one...
{ "chunk_text": "Chunk 1: hello, my name is emily. i'm looking for a place to stay.\nChunk 2: i'm looking for, uh, for one which is, uh, in the expensive price inch.\nChunk 3: okay, i'm looking at one which is just below a three star rating.\nChunk 4: okay, i'm looking for one which is not a guest house in the east a...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello, my name is emily. i'm looking for a place to stay.", "role": "user", "streaming_reasoning": "The user mentions Hello, my name is emily. i'm looking for a place to...
{ "answer": "Sure. I can provide the postcode, internet for the matching hotel option.", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: hello, my name is emily; i'm looking for a place to stay; i'm looking for, uh, for one which is, uh, in the expensive price inch; okay, i'...
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3446", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_sdeval_v02_000161
train
audio
multi_turn
[ { "role": "user", "text": "Ambrose met me at the bottom of the stairs and showed me the way to the supper room.", "turn_id": 1 } ]
{ "chunk_text": "Chunk 1: Ambrose met me at the bottom of the stairs and showed me the way to the supper room.", "dialogue_text": "Chunk 1: Ambrose met me at the bottom of the stairs and showed me the way to the supper room.", "instruction": "Help the user complete a spoken task-oriented dialogue based on gradual...
{ "num_checkpoints": 1, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Ambrose met me at the bottom of the stairs and showed me the way to the supper room.", "role": "user", "streaming_reasoning": "The user mentions talking about Ambrose. I...
{ "answer": "It sounds like a perfect night for a warm supper. Did Ambrose seem at ease, despite the storm outside?", "reasoning": "Across the dialogue, the user discusses \"Ambrose met me at the bottom of the stairs and showed me.\" over 1 turns. The answer should respond to the latest user turn naturally, maintai...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "easy", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SD-Eval", "dialogue_id": "5142-36377-0003-raining", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004479
train
audio
multi_turn
[ { "role": "user", "text": "i.", "turn_id": 1 }, { "role": "assistant", "text": "n n.", "turn_id": 2 }, { "role": "user", "text": "yes. uh, the spelling is j.", "turn_id": 3 }, { "role": "assistant", "text": "ah.", "turn_id": 4 }, { "role": "user", ...
{ "chunk_text": "Chunk 1: i.\nChunk 2: n n.\nChunk 3: yes. uh, the spelling is j.\nChunk 4: ah.\nChunk 5: w i.\nChunk 6: inset.\nChunk 7: on the.\nChunk 8: add up the dot com.", "dialogue_text": "Chunk 1: i.\nChunk 2: n n.\nChunk 3: yes. uh, the spelling is j.\nChunk 4: ah.\nChunk 5: w i.\nChunk 6: inset.\nChunk 7:...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i.", "role": "user", "streaming_reasoning": "The user mentions I. It is not yet clear whether the user needs the full scope of the user's request. The assistant should a...
{ "answer": "I can help. What task or place should I focus on?", "reasoning": "Across the dialogue, the user is talking about uh, the spelling is j. Key context: uh, the spelling is j; add up the dot com. The answer should focus on the latest user turn while incorporating relevant context from earlier turns.", "r...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1961", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004480
train
audio
multi_turn
[ { "role": "user", "text": "oh, i like to go there at a five o'clock p. m..", "turn_id": 1 }, { "role": "assistant", "text": "how about just four o'clock p. m..", "turn_id": 2 }, { "role": "user", "text": "okay, my name is hector seven.", "turn_id": 3 }, { "role": ...
{ "chunk_text": "Chunk 1: oh, i like to go there at a five o'clock p. m..\nChunk 2: how about just four o'clock p. m..\nChunk 3: okay, my name is hector seven.\nChunk 4: it's hectorhector and severnseven.\nChunk 5: yes, i'm looking for a place to go in town. do you have any recommendation.\nChunk 6: i like it to stay...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "oh, i like to go there at a five o'clock p. m..", "role": "user", "streaming_reasoning": "The user mentions Oh, i like to go there at a five o'clock p. m. It is not yet ...
{ "answer": "Sure. I can provide the address for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for place to go in town. Key context: oh, i like to go there at a five o'clock p; how about just four o'clock p; okay, my name is hector seven; it's hectorhector and severnseve...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2919", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_sdeval_v02_000162
train
audio
multi_turn
[ { "role": "user", "text": "How can I relax after a stressful day?", "turn_id": 1 } ]
{ "chunk_text": "Chunk 1: How can I relax after a stressful day?", "dialogue_text": "Chunk 1: How can I relax after a stressful day?", "instruction": "Help the user complete a spoken task-oriented dialogue based on gradually revealed audio/transcript chunks.", "length_bucket": "short", "num_turns": 1, "tota...
{ "num_checkpoints": 1, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "How can I relax after a stressful day?", "role": "user", "streaming_reasoning": "The user mentions How can I relax after a stressful day?. It is not yet clear whether th...
{ "answer": "A walk along the beach could be incredibly soothing; the sound of the waves has a naturally calming effect.", "reasoning": "Across the dialogue, the user discusses \"How can I relax after a stressful.\" over 1 turns. The answer should respond to the latest user turn naturally, maintaining the conversat...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "easy", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SD-Eval", "dialogue_id": "185_0", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004481
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "hi, i'm looking for a vietnamese restaurant in the city centre.", "turn_id": 2 }, { "role": "user", "text": "okay. um, what about african, i think african restaurant.", "turn_id": 3 }, ...
{ "chunk_text": "Chunk 1: hello.\nChunk 2: hi, i'm looking for a vietnamese restaurant in the city centre.\nChunk 3: okay. um, what about african, i think african restaurant.\nChunk 4: okay. can i have the address for that place.\nChunk 5: alright, and i'm also looking for a particular hotel called kirkwood house.\nC...
{ "num_checkpoints": 6, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "I can help with the hotel. Could you share the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for vietnamese restaurant in the city centre. Key context: hi, i'm looking for a vietnamese restaurant in the city centre; um, what about african, i think african restauran...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "opinion_sharing" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3237", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004482
train
audio
multi_turn
[ { "role": "user", "text": "i'm good. how are you.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for a train.", "turn_id": 2 }, { "role": "user", "text": "benningham new street.", "turn_id": 3 }, { "role": "assistant", "text": "i should go to ca...
{ "chunk_text": "Chunk 1: i'm good. how are you.\nChunk 2: i'm looking for a train.\nChunk 3: benningham new street.\nChunk 4: i should go to cambridge.\nChunk 5: um if it could arrive by half past 6.\nChunk 6: the day before sunday.\nChunk 7: okay, that's perfect. can i make a booking.\nChunk 8: that is. 8, 8, 8.", ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i'm good. how are you.", "role": "user", "streaming_reasoning": "The user mentions I'm good. how are you. It is not yet clear whether the user needs the full scope of th...
{ "answer": "I can help book that with the details you gave. Would you like me to confirm the reservation?", "reasoning": "Across the dialogue, the user is searching for train. Key context: how are you; i'm looking for a train; benningham new street; i should go to cambridge. The answer should focus on the latest u...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1167", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001318
train
audio
multi_turn
[ { "role": "user", "text": "Human: Sounds like you know a lot about that prophecy, sounds interesing", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I am just an acolyte, not a true follower of the NFL prophecy, which is given to many but understood by few. I hope that next Super...
{ "chunk_text": "Chunk 1: Human: Sounds like you know a lot about that prophecy, sounds interesing\nChunk 2: Assistant: I am just an acolyte, not a true follower of the NFL prophecy, which is given to many but understood by few. I hope that next Super Bowl shows us the truth.\nChunk 3: Human: I hope it does I cant st...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Sounds like you know a lot about that prophecy, sounds interesing", "role": "user", "streaming_reasoning": "The user mentions talking about Human, Sounds. It is n...
{ "answer": "The prophecy will be fulfilled in its time, whether we hope or not.", "reasoning": "Across the dialogue, the user is asking for help with human: sounds like you know a lot about that prophecy, sounds interesing assistant: i am just an acolyte, not a true follower of the nfl prophecy, which is given to ...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_6bf2033d-1e44-4edd-a14d-f468dff6f526", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004483
train
audio
multi_turn
[ { "role": "user", "text": "mhm. okay, sounds good. so can you make a reservation for me.", "turn_id": 1 }, { "role": "assistant", "text": "yeah, sure. um my full name is shirley grass. s h irl ey for shirley.", "turn_id": 2 }, { "role": "user", "text": "and gr o double f for ...
{ "chunk_text": "Chunk 1: mhm. okay, sounds good. so can you make a reservation for me.\nChunk 2: yeah, sure. um my full name is shirley grass. s h irl ey for shirley.\nChunk 3: and gr o double f for ground.\nChunk 4: um can you just text me.\nChunk 5: yeah, uh, okay. by the way, uh, we have 6 people, and we'd like t...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "mhm. okay, sounds good. so can you make a reservation for me.", "role": "user", "streaming_reasoning": "The user mentions Mhm. It is not yet clear whether the user needs...
{ "answer": "I can help with the train. Could you share the day, time?", "reasoning": "Across the dialogue, the user is searching for train. Key context: okay, sounds good; so can you make a reservation for me; um my full name is shirley grass; s h irl ey for shirley. The answer should focus on the latest user turn...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "hard", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0110", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004484
train
audio
multi_turn
[ { "role": "user", "text": "thank you very much and i also want to book a taxi to commute between the two policies.", "turn_id": 1 }, { "role": "assistant", "text": "uh, i want to leave attraction by one hour before 6 am.", "turn_id": 2 }, { "role": "user", "text": "yes.", ...
{ "chunk_text": "Chunk 1: thank you very much and i also want to book a taxi to commute between the two policies.\nChunk 2: uh, i want to leave attraction by one hour before 6 am.\nChunk 3: yes.\nChunk 4: uh, i want to the phone number of the taxi.\nChunk 5: okay, sure. my phone number is 9 3 8 9.\nChunk 6: and a dou...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "thank you very much and i also want to book a taxi to commute between the two policies.", "role": "user", "streaming_reasoning": "The user mentions wanting to book Thank...
{ "answer": "Sure. I can provide the phone for the matching taxi option.", "reasoning": "Across the dialogue, the user is wanting to book taxi to commute between the two policies. Key context: thank you very much and i also want to book a taxi to commute between the two policies; uh, i want to leave attraction by o...
{ "category": "finance_consumer", "subcategory": "budget_planning", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0684", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004485
train
audio
multi_turn
[ { "role": "user", "text": "m i r iam is the first name.", "turn_id": 1 }, { "role": "assistant", "text": "th oma s is the last man.", "turn_id": 2 }, { "role": "user", "text": "yes, i will still add up for you.", "turn_id": 3 }, { "role": "assistant", "text": ...
{ "chunk_text": "Chunk 1: m i r iam is the first name.\nChunk 2: th oma s is the last man.\nChunk 3: yes, i will still add up for you.\nChunk 4: m t. af.\nChunk 5: 3 w 9 r.\nChunk 6: outlook dot com.\nChunk 7: i'm also looking for trains.\nChunk 8: for. it would have to depart from cambridge.", "dialogue_text": "Ch...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "m i r iam is the first name.", "role": "user", "streaming_reasoning": "The user mentions M i r iam is the first name. It is not yet clear whether the user needs the full...
{ "answer": "I can help with the train. Could you share the destination, day, time?", "reasoning": "Across the dialogue, the user is searching for trains. Key context: m i r iam is the first name; th oma s is the last man; yes, i will still add up for you; outlook dot com. The answer should focus on the latest user...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3119", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001319
train
audio
multi_turn
[ { "role": "user", "text": "Human: Well, artists need to be paid. I thought she had the disagreement with Apple Co. not spotify.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Probably both lol... I agree with her but why not advertise your music on the radio?", "turn_id": 2...
{ "chunk_text": "Chunk 1: Human: Well, artists need to be paid. I thought she had the disagreement with Apple Co. not spotify.\nChunk 2: Assistant: Probably both lol... I agree with her but why not advertise your music on the radio?\nChunk 3: Human: Why the radio? How about social media or you tube instead?", "dial...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Well, artists need to be paid. I thought she had the disagreement with Apple Co. not spotify.", "role": "user", "streaming_reasoning": "The user mentions talking ...
{ "answer": "True, but I am not sure how she advertises but spotify is just a radio app.", "reasoning": "Across the dialogue, the user discusses \"Well, artists need to be paid. I thought she had the.\" over 3 turns. The answer should respond to the latest user turn naturally, maintaining the conversational tone.",...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "explain_cause" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_4190e346-b794-45e7-a8d9-fa43b4016790", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001320
train
audio
multi_turn
[ { "role": "user", "text": "Human: Im guessing its because of that, engineers tend to have longer careers also so yeah, that makes sense.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Did you know the av. NFL game has only 11 minutes of live game play? Guess they know: have emb...
{ "chunk_text": "Chunk 1: Human: Im guessing its because of that, engineers tend to have longer careers also so yeah, that makes sense.\nChunk 2: Assistant: Did you know the av. NFL game has only 11 minutes of live game play? Guess they know: have embedded chips in players' shoulder pads these days. Do they really ne...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Im guessing its because of that, engineers tend to have longer careers also so yeah, that makes sense.", "role": "user", "streaming_reasoning": "The user mentions...
{ "answer": "Chip no problem for my fav. player, RB Saquon Barkley of Giants. Makes you want to watch football all day - great young athlete.", "reasoning": "Across the dialogue, the user is asking for help with to track every move w/cam human: only 11 minutes really. Key context: Human: Im guessing its because of ...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "hard", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_58d231e0-66fd-4615-a61f-4c4a8173ee52", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004486
train
audio
multi_turn
[ { "role": "user", "text": "god, thank you very much.", "turn_id": 1 }, { "role": "assistant", "text": "yes, i'm also looking for a trend.", "turn_id": 2 }, { "role": "user", "text": "uh, i think the trend should leave after approach to 5 p.", "turn_id": 3 }, { "ro...
{ "chunk_text": "Chunk 1: god, thank you very much.\nChunk 2: yes, i'm also looking for a trend.\nChunk 3: uh, i think the trend should leave after approach to 5 p.\nChunk 4: oh, yes, that's right. and. and should go to where i'm now and i'm now in the bishop, stockport.\nChunk 5: that's right. and it should live on ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "god, thank you very much.", "role": "user", "streaming_reasoning": "The user mentions God, thank you very much. It is not yet clear whether the user needs the full scope...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is searching for trend. Key context: god, thank you very much; yes, i'm also looking for a trend; uh, i think the trend should leave after approach to 5 p; oh, yes, that's right. The answer should ...
{ "category": "tech_support", "subcategory": "device_troubleshooting", "difficulty": "medium", "intent_type": "opinion_sharing" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1127", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004487
train
audio
multi_turn
[ { "role": "user", "text": "what. i'm looking for a dream, please.", "turn_id": 1 }, { "role": "assistant", "text": "i should leave after 20 o'clock.", "turn_id": 2 }, { "role": "user", "text": "and it should leave on wednesday.", "turn_id": 3 }, { "role": "assista...
{ "chunk_text": "Chunk 1: what. i'm looking for a dream, please.\nChunk 2: i should leave after 20 o'clock.\nChunk 3: and it should leave on wednesday.\nChunk 4: yes, and to depart from bishop stafford.\nChunk 5: yes, and should go to cambridge.\nChunk 6: yes, that's all my comments.\nChunk 7: okay, thank you.\nChunk...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "what. i'm looking for a dream, please.", "role": "user", "streaming_reasoning": "The user mentions What. i'm looking for a dream, please. It is not yet clear whether the...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is searching for dream. Key context: i'm looking for a dream, please; i should leave after 20 o'clock; and it should leave on wednesday; yes, and to depart from bishop stafford. The answer should f...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3930", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004488
train
audio
multi_turn
[ { "role": "user", "text": "so basically, i'm looking for places to go to in town. the attraction should be in the east.", "turn_id": 1 }, { "role": "assistant", "text": "ok.", "turn_id": 2 }, { "role": "user", "text": "okay. alright. and what type of an attraction is it.", ...
{ "chunk_text": "Chunk 1: so basically, i'm looking for places to go to in town. the attraction should be in the east.\nChunk 2: ok.\nChunk 3: okay. alright. and what type of an attraction is it.\nChunk 4: okay. do you do do you have an entrance fee.\nChunk 5: and what is their opening hours.\nChunk 6: okay, alright,...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "so basically, i'm looking for places to go to in town. the attraction should be in the east.", "role": "user", "streaming_reasoning": "The user mentions So basically, i'...
{ "answer": "I can help with the hotel. Could you share the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for places to go to in town. Key context: so basically, i'm looking for places to go to in town; the attraction should be in the east; and what type of an attraction is it;...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1602", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001321
train
audio
multi_turn
[ { "role": "user", "text": "Human: Hello there, are you a big baseball fan?", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Yes I am, I like the Yankees what about you?", "turn_id": 2 }, { "role": "user", "text": "Human: That's awesome, yeah I am too, but I am a B...
{ "chunk_text": "Chunk 1: Human: Hello there, are you a big baseball fan?\nChunk 2: Assistant: Yes I am, I like the Yankees what about you?\nChunk 3: Human: That's awesome, yeah I am too, but I am a Brewers fan, we almost made the world series this year!", "dialogue_text": "Chunk 1: Human: Hello there, are you a bi...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Hello there, are you a big baseball fan?", "role": "user", "streaming_reasoning": "The user mentions Human: Hello there, are you a big baseball fan?. It is not ye...
{ "answer": "Oh thats too bad! hopefully next year you can make it, would be nice to see a defining match between the Yankees and the Brewers.", "reasoning": "Across the dialogue, the user discusses \"Hello there, are you a big baseball.\" over 3 turns. The answer should respond to the latest user turn naturally, m...
{ "category": "information_extraction", "subcategory": "constraints_extraction", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_1271586a-38c8-4e7d-bfa7-660aaab4e8d9", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004489
train
audio
multi_turn
[ { "role": "user", "text": "no, 1015, 1015, sorry.", "turn_id": 1 }, { "role": "assistant", "text": "got a passing.", "turn_id": 2 }, { "role": "user", "text": "the 8 43 will do you compare.", "turn_id": 3 }, { "role": "assistant", "text": "the 8 43.", "tur...
{ "chunk_text": "Chunk 1: no, 1015, 1015, sorry.\nChunk 2: got a passing.\nChunk 3: the 8 43 will do you compare.\nChunk 4: the 8 43.\nChunk 5: um what time will it leave and the travel time.\nChunk 6: okay. can i just get the train id.\nChunk 7: okay, thank you so much.\nChunk 8: yes, please. i'm also looking for a ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "no, 1015, 1015, sorry.", "role": "user", "streaming_reasoning": "The user mentions No, 1015, 1015, sorry. It is not yet clear whether the user needs the full scope of th...
{ "answer": "I can help with the hotel. Could you share the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: no, 1015, 1015, sorry; got a passing; the 8 43 will do you compare; um what time will it leave and the travel time. The answer should focus ...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1443", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004490
train
audio
multi_turn
[ { "role": "user", "text": "thank you so much.", "turn_id": 1 }, { "role": "assistant", "text": "great. could you please make a booking for me.", "turn_id": 2 }, { "role": "user", "text": "um we have 6, 7 people in total.", "turn_id": 3 }, { "role": "assistant", ...
{ "chunk_text": "Chunk 1: thank you so much.\nChunk 2: great. could you please make a booking for me.\nChunk 3: um we have 6, 7 people in total.\nChunk 4: yes, of course. my id number is a 6 4 3 1.\nChunk 5: uh, 9 6 0.\nChunk 6: 3 0, 0, 2, 4.\nChunk 7: and 7 1 1 6.\nChunk 8: i think this will be all thank you.", "d...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "thank you so much.", "role": "user", "streaming_reasoning": "The user mentions Thank you so much. It is not yet clear whether the user needs the full scope of the user's...
{ "answer": "I can help with the train. Could you share the departure, destination, day?", "reasoning": "Across the dialogue, the user is wanting to book thank you so much. Key context: thank you so much; could you please make a booking for me; um we have 6, 7 people in total; yes, of course. The answer should focu...
{ "category": "education", "subcategory": "course_planning", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0440", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004491
train
audio
multi_turn
[ { "role": "user", "text": "yes. it is called the traveler's threat.", "turn_id": 1 }, { "role": "assistant", "text": "sure, please make a resolution for me.", "turn_id": 2 }, { "role": "user", "text": "hmm, it will be totally.", "turn_id": 3 }, { "role": "assistan...
{ "chunk_text": "Chunk 1: yes. it is called the traveler's threat.\nChunk 2: sure, please make a resolution for me.\nChunk 3: hmm, it will be totally.\nChunk 4: um half past 11 o'clock.\nChunk 5: um tomorrow.\nChunk 6: bright.\nChunk 7: hmm, sure. my name. chinese john.\nChunk 8: c h e l. e i f i e.", "dialogue_tex...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes. it is called the traveler's threat.", "role": "user", "streaming_reasoning": "The user mentions Yes. it is called the traveler's threat. It is not yet clear whether...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is talking about the traveler's threat. Key context: it is called the traveler's threat; sure, please make a resolution for me; hmm, it will be totally; um half past 11 o'clock. The answ...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0173", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004492
train
audio
multi_turn
[ { "role": "user", "text": "hi this is james speaking and i am looking for some information on a particular restaurant called shiraz are you able to assist me with this", "turn_id": 1 }, { "role": "assistant", "text": "okay. um, could you please look for a table for myself and my partner and ...
{ "chunk_text": "Chunk 1: hi this is james speaking and i am looking for some information on a particular restaurant called shiraz are you able to assist me with this\nChunk 2: okay. um, could you please look for a table for myself and my partner and four other people tuesday at one pm.\nChunk 3: okay. how about maki...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi this is james speaking and i am looking for some information on a particular restaurant called shiraz are you able to assist me with this", "role": "user", "streaming...
{ "answer": "I can help book that. Could you provide the area, food, price?", "reasoning": "Across the dialogue, the user is searching for packing spot for myself. Key context: hi this is james speaking and i am looking for some information on a particular restaurant called shiraz are you able to assist me with thi...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "hard", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1176", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001322
train
audio
multi_turn
[ { "role": "user", "text": "Human: And yet somehow their music has been timeless, it was even the best selling album of the 2000's!", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: That is consistent with their book being fantastic. My wife is a talented musician and says that alm...
{ "chunk_text": "Chunk 1: Human: And yet somehow their music has been timeless, it was even the best selling album of the 2000's!\nChunk 2: Assistant: That is consistent with their book being fantastic. My wife is a talented musician and says that almost all of the songs have been done better by someone else than by ...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: And yet somehow their music has been timeless, it was even the best selling album of the 2000's!", "role": "user", "streaming_reasoning": "The user mentions askin...
{ "answer": "I hadn't heard that. I wonder how Lennon and McCartney wrote their songs?", "reasoning": "Across the dialogue, the user is wanting to book being fantastic. Key context: Human: And yet somehow their music has been timeless, it was even the best selling album of the 2000's; Assistant: That is consistent ...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "hard", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_3b5526d9-5ff3-4698-9f21-022a3c0103ff", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004493
train
audio
multi_turn
[ { "role": "user", "text": "okay.", "turn_id": 1 }, { "role": "assistant", "text": "sure, my name is michelle dego.", "turn_id": 2 }, { "role": "user", "text": "shell i m i c h.", "turn_id": 3 }, { "role": "assistant", "text": "e l n e.", "turn_id": 4 }, ...
{ "chunk_text": "Chunk 1: okay.\nChunk 2: sure, my name is michelle dego.\nChunk 3: shell i m i c h.\nChunk 4: e l n e.\nChunk 5: an eagle di.\nChunk 6: yes, that's fine.\nChunk 7: m d a i.\nChunk 8: gli.", "dialogue_text": "Chunk 1: okay.\nChunk 2: sure, my name is michelle dego.\nChunk 3: shell i m i c h.\nChunk ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay.", "role": "user", "streaming_reasoning": "The user mentions Okay. It is not yet clear whether the user needs the full scope of the user's request. The assistant sh...
{ "answer": "I can help. What task or place should I focus on?", "reasoning": "Across the dialogue, the user is talking about sure, my name is michelle dego. Key context: sure, my name is michelle dego; shell i m i c h; an eagle di; yes, that's fine. The answer should focus on the latest user turn while incorporati...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2065", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001323
train
audio
multi_turn
[ { "role": "user", "text": "Human: For sure, did you know in high school though, Tom Bradys JV team never won a game and the coach refused to put him in!", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: HAHA! Well he is probably regretting that decision!", "turn_id": 2 }, ...
{ "chunk_text": "Chunk 1: Human: For sure, did you know in high school though, Tom Bradys JV team never won a game and the coach refused to put him in!\nChunk 2: Assistant: HAHA! Well he is probably regretting that decision!\nChunk 3: Human: I bet lol. Do you follow College football?", "dialogue_text": "Chunk 1: Hu...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: For sure, did you know in high school though, Tom Bradys JV team never won a game and the coach refused to put him in!", "role": "user", "streaming_reasoning": "T...
{ "answer": "Not very much. My husband does though, so we watch UGA play a lot and he likes to watch Sony Michel with the Patriots.", "reasoning": "Across the dialogue, the user discusses \"For sure, did you know in high school though, Tom Bradys JV.\" over 3 turns. The answer should respond to the latest user turn...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_0cbf608e-e8f5-4108-9cf0-a3338a42fd4b", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004494
train
audio
multi_turn
[ { "role": "user", "text": "morning. my name is joella. i'm looking for a place to dine in cambridge.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for an expensive place that serves. let's say, indian food in cambridge.", "turn_id": 2 }, { "role": "user", "tex...
{ "chunk_text": "Chunk 1: morning. my name is joella. i'm looking for a place to dine in cambridge.\nChunk 2: i'm looking for an expensive place that serves. let's say, indian food in cambridge.\nChunk 3: sure things.\nChunk 4: okay, excellent. can i please have their post code.\nChunk 5: and in which area is it loca...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "morning. my name is joella. i'm looking for a place to dine in cambridge.", "role": "user", "streaming_reasoning": "The user mentions searching for Morning. It is not ye...
{ "answer": "Sure. I can provide the postcode, area, address for the matching restaurant option.", "reasoning": "Across the dialogue, the user is searching for place to dine in cambridge. Key context: my name is joella; i'm looking for a place to dine in cambridge; i'm looking for an expensive place that serves; le...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG1569", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004495
train
audio
multi_turn
[ { "role": "user", "text": "ong.", "turn_id": 1 }, { "role": "assistant", "text": "okay, and so need train information.", "turn_id": 2 }, { "role": "user", "text": "um i'd like to take the train to catch same reservation.", "turn_id": 3 }, { "role": "assistant", ...
{ "chunk_text": "Chunk 1: ong.\nChunk 2: okay, and so need train information.\nChunk 3: um i'd like to take the train to catch same reservation.\nChunk 4: departing from norwich.\nChunk 5: okay, i. okay, i need i want to arrive in cambridge by nine. do you not want to do that.\nChunk 6: okay. um. would you know what ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "ong.", "role": "user", "streaming_reasoning": "The user mentions Ong. It is not yet clear whether the user needs the full scope of the user's request. The assistant shou...
{ "answer": "Sure. I can provide the travel_time for the matching train option.", "reasoning": "Across the dialogue, the user is talking about train information. Key context: okay, and so need train information; um i'd like to take the train to catch same reservation; departing from norwich; okay, i need i want to ...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2356", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004496
train
audio
multi_turn
[ { "role": "user", "text": "and my last name is g a r d a.", "turn_id": 1 }, { "role": "assistant", "text": "uh, no, that's not necessary.", "turn_id": 2 }, { "role": "user", "text": "uh, also i'm looking for a train.", "turn_id": 3 }, { "role": "assistant", "t...
{ "chunk_text": "Chunk 1: and my last name is g a r d a.\nChunk 2: uh, no, that's not necessary.\nChunk 3: uh, also i'm looking for a train.\nChunk 4: uh, we are at cambridge right now and we plan to go to standard airport.\nChunk 5: uh, we want the train to leave after 1145 am.\nChunk 6: on today's sunday. so tomorr...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "and my last name is g a r d a.", "role": "user", "streaming_reasoning": "The user mentions And my last name is g a r d a. It is not yet clear whether the user needs the ...
{ "answer": "Sure. I can provide the id, ticket for the matching train option.", "reasoning": "Across the dialogue, the user is searching for train. Key context: and my last name is g a r d a; uh, no, that's not necessary; uh, also i'm looking for a train; uh, we are at cambridge right now and we plan to go to stan...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1100", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004497
train
audio
multi_turn
[ { "role": "user", "text": "to the. good day.", "turn_id": 1 }, { "role": "assistant", "text": "um could you please assist me. i am looking for a place to stay, please.", "turn_id": 2 }, { "role": "user", "text": "so i should be in the north and include free wifi. and the hote...
{ "chunk_text": "Chunk 1: to the. good day.\nChunk 2: um could you please assist me. i am looking for a place to stay, please.\nChunk 3: so i should be in the north and include free wifi. and the hotel should be in the star rating of four stars.", "dialogue_text": "Chunk 1: to the. good day.\nChunk 2: um could you ...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "to the. good day.", "role": "user", "streaming_reasoning": "The user mentions To the. good day. It is not yet clear whether the user needs the full scope of the user's r...
{ "answer": "I can help with the hotel. Could you share the price, hotel_type?", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: um could you please assist me; i am looking for a place to stay, please; so i should be in the north and include free wifi; and the hotel should b...
{ "category": "tech_support", "subcategory": "device_troubleshooting", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2457", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004498
train
audio
multi_turn
[ { "role": "user", "text": "is that perfect. and what's the price for this.", "turn_id": 1 }, { "role": "assistant", "text": "yes.", "turn_id": 2 }, { "role": "user", "text": "okay and tell me something i'm also looking for a recommendation for a hotel it doesn't need to have ...
{ "chunk_text": "Chunk 1: is that perfect. and what's the price for this.\nChunk 2: yes.\nChunk 3: okay and tell me something i'm also looking for a recommendation for a hotel it doesn't need to have free parking uh but it needs\nChunk 4: to be expensive i'm looking for something really nice\nChunk 5: okay, thank you...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "is that perfect. and what's the price for this.", "role": "user", "streaming_reasoning": "The user mentions Is that perfect. and what's the price for this. It is not yet...
{ "answer": "Sure. I can provide the postcode for the matching hotel option.", "reasoning": "Across the dialogue, the user is searching for recommendation for a hotel it doesn't need to have free parking uh. Key context: is that perfect; and what's the price for this; okay and tell me something i'm also looking for...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "hard", "intent_type": "fact_lookup" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2688", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001324
train
audio
multi_turn
[ { "role": "user", "text": "Human: Maybe. The prime minister of uk has a cat, and it has the title of chief mouser, funny huh", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Chief mouser? Tupac was mouse king in the nutcracker, I don't think they would go well together.", "tu...
{ "chunk_text": "Chunk 1: Human: Maybe. The prime minister of uk has a cat, and it has the title of chief mouser, funny huh\nChunk 2: Assistant: Chief mouser? Tupac was mouse king in the nutcracker, I don't think they would go well together.\nChunk 3: Human: In Japan, the prime minister sends a silver cup to celebrat...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: Maybe. The prime minister of uk has a cat, and it has the title of chief mouser, funny huh", "role": "user", "streaming_reasoning": "The user mentions talking abo...
{ "answer": "They are ceremonial, they don't need to be sturdy, not like pointes for ballet dancers, they can go through 4 pairs in a week.", "reasoning": "Across the dialogue, the user discusses \"Maybe. The prime minister of uk has a cat, and it has the.\" over 3 turns. The answer should respond to the latest use...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_374ae28b-9bc0-45f8-ab39-a52c164c8e79", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004499
train
audio
multi_turn
[ { "role": "user", "text": "hi, good morning. this is kevin.", "turn_id": 1 }, { "role": "assistant", "text": "i'm looking for places to go in town. do you have any recommendation of great colleagues.", "turn_id": 2 }, { "role": "user", "text": "yeah, the center room will be g...
{ "chunk_text": "Chunk 1: hi, good morning. this is kevin.\nChunk 2: i'm looking for places to go in town. do you have any recommendation of great colleagues.\nChunk 3: yeah, the center room will be great.\nChunk 4: cherish college, great. what's the post code.\nChunk 5: c b 23 b u okay, got it, thank you.\nChunk 6: ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, good morning. this is kevin.", "role": "user", "streaming_reasoning": "The user mentions Hi, good morning. this is kevin. It is not yet clear whether the user needs ...
{ "answer": "Sure. I can provide the postcode for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for places to go in town. Key context: hi, good morning; this is kevin; i'm looking for places to go in town; do you have any recommendation of great colleagues. The answer sh...
{ "category": "education", "subcategory": "course_planning", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0231", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004500
train
audio
multi_turn
[ { "role": "user", "text": "uh, can you check for me if there is a restaurant that serves the asian.", "turn_id": 1 }, { "role": "assistant", "text": "yeah.", "turn_id": 2 }, { "role": "user", "text": "i would like to make a.", "turn_id": 3 }, { "role": "assistant"...
{ "chunk_text": "Chunk 1: uh, can you check for me if there is a restaurant that serves the asian.\nChunk 2: yeah.\nChunk 3: i would like to make a.\nChunk 4: for myself and five times.\nChunk 5: better than.\nChunk 6: at 4 pm.\nChunk 7: okay.\nChunk 8: yes, i'm sorry for a place. ya.", "dialogue_text": "Chunk 1: u...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "uh, can you check for me if there is a restaurant that serves the asian.", "role": "user", "streaming_reasoning": "The user mentions Uh, can you check for me if there is...
{ "answer": "I can help with the attraction. Could you share the area, type?", "reasoning": "Across the dialogue, the user is talking about to make a. Key context: uh, can you check for me if there is a restaurant that serves the asian; i would like to make a; for myself and five times; better than. The answer shou...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2742", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004501
train
audio
multi_turn
[ { "role": "user", "text": "no, can i leave after copper 5 p.", "turn_id": 1 }, { "role": "assistant", "text": "yes, no problem. i'm in cambridge.", "turn_id": 2 }, { "role": "user", "text": "sure i will.", "turn_id": 3 }, { "role": "assistant", "text": "not a ...
{ "chunk_text": "Chunk 1: no, can i leave after copper 5 p.\nChunk 2: yes, no problem. i'm in cambridge.\nChunk 3: sure i will.\nChunk 4: not a problem.\nChunk 5: excellent, please look at for two adults.\nChunk 6: my id is 1 3. 1 9 1.\nChunk 7: 2 3, 7 2 6.\nChunk 8: 7 5 9 2. 5 4 did you get.", "dialogue_text": "Ch...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "no, can i leave after copper 5 p.", "role": "user", "streaming_reasoning": "The user mentions No, can i leave after copper 5 p. It is not yet clear whether the user need...
{ "answer": "I can help with the train. Could you share the destination, day?", "reasoning": "Across the dialogue, the user is reporting a problem with no, can i leave after copper 5 p. Key context: no, can i leave after copper 5 p; yes, no problem; i'm in cambridge; sure i will. The answer should focus on the late...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2828", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004502
train
audio
multi_turn
[ { "role": "user", "text": "hi, how are you doing today.", "turn_id": 1 }, { "role": "assistant", "text": "so i'm not. came for a place to stay in cambridge. can you help me with that.", "turn_id": 2 } ]
{ "chunk_text": "Chunk 1: hi, how are you doing today.\nChunk 2: so i'm not. came for a place to stay in cambridge. can you help me with that.", "dialogue_text": "Chunk 1: hi, how are you doing today.\nChunk 2: so i'm not. came for a place to stay in cambridge. can you help me with that.", "instruction": "Help th...
{ "num_checkpoints": 2, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hi, how are you doing today.", "role": "user", "streaming_reasoning": "The user mentions Hi, how are you doing today. It is not yet clear whether the user needs the full...
{ "answer": "I can help with the hotel. Could you share the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is asking for help with hi, how are you doing today. Key context: hi, how are you doing today; came for a place to stay in cambridge; can you help me with that. The answer should focus ...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1807", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001325
train
audio
multi_turn
[ { "role": "user", "text": "Human: I guess they arent religious, they sure lost a lot of votes because of that,", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: I wouldn't think so, as people back then knew how to separate country and religion. Unlike today. Weird though, they did...
{ "chunk_text": "Chunk 1: Human: I guess they arent religious, they sure lost a lot of votes because of that,\nChunk 2: Assistant: I wouldn't think so, as people back then knew how to separate country and religion. Unlike today. Weird though, they did include passages from the bible in schools.\nChunk 3: Human: I thi...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: I guess they arent religious, they sure lost a lot of votes because of that,", "role": "user", "streaming_reasoning": "The user mentions talking about Human. It i...
{ "answer": "I don't think so either. I do find it ineresting that apple means both fruit and malace.", "reasoning": "The dialogue is about \"I guess they arent religious, they sure lost a lot of votes.\" The user later explains \"I think religion shouldnt be mandatory we all have.\" The answer should treat this as...
{ "category": "information_extraction", "subcategory": "long_note_to_summary", "difficulty": "medium", "intent_type": "opinion_sharing" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_0286e13f-801d-48cd-bfec-1dcd61f11aa9", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004503
train
audio
multi_turn
[ { "role": "user", "text": "thank you. can you also give me the train id as well.", "turn_id": 1 }, { "role": "assistant", "text": "thank you for that i'm also looking for information on attractions that i can visit in the town is this something that you would also", "turn_id": 2 }, {...
{ "chunk_text": "Chunk 1: thank you. can you also give me the train id as well.\nChunk 2: thank you for that i'm also looking for information on attractions that i can visit in the town is this something that you would also\nChunk 3: be able to help\nChunk 4: okay so the attraction that i'm looking for should be loca...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "thank you. can you also give me the train id as well.", "role": "user", "streaming_reasoning": "The user mentions Thank you. can you also give me the train id as well. I...
{ "answer": "Sure. I can provide the fee for the matching attraction option.", "reasoning": "Across the dialogue, the user is searching for can you also give me the train id as well. Key context: can you also give me the train id as well; how about one that is in the type of a college; okay, thank you so much for t...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "hard", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3670", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001326
train
audio
multi_turn
[ { "role": "user", "text": "Human: If they are good enough then so be it. Maybe she can then write a book and make a documentary about how we should live.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: Tom Brady's documentary was on Facebook watch, I doubt many people saw, she s...
{ "chunk_text": "Chunk 1: Human: If they are good enough then so be it. Maybe she can then write a book and make a documentary about how we should live.\nChunk 2: Assistant: Tom Brady's documentary was on Facebook watch, I doubt many people saw, she should not make the same mistake.\nChunk 3: Human: The first female ...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: If they are good enough then so be it. Maybe she can then write a book and make a documentary about how we should live.", "role": "user", "streaming_reasoning": "...
{ "answer": "But I don't think she will be as popular as Tom Brady around Boston, he is seen as a messiah there.", "reasoning": "Across the dialogue, the user is wanting to book and make a documentary about how we should live. Key context: Human: If they are good enough then so be it; Maybe she can then write a boo...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_96d4d5ac-d651-4fc6-b29c-31d9cfba8247", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004504
train
audio
multi_turn
[ { "role": "user", "text": "ok, thank you so much.", "turn_id": 1 }, { "role": "assistant", "text": "yes, that will be fine.", "turn_id": 2 }, { "role": "user", "text": "okay, sound correct. this can make working for me one person as well.", "turn_id": 3 }, { "role...
{ "chunk_text": "Chunk 1: ok, thank you so much.\nChunk 2: yes, that will be fine.\nChunk 3: okay, sound correct. this can make working for me one person as well.\nChunk 4: uh, okay. my i number is 2 9 3.\nChunk 5: 4, double 7 6.\nChunk 6: 4 7 double 2.\nChunk 7: one double pm double d.\nChunk 8: okay, that sounds pe...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "ok, thank you so much.", "role": "user", "streaming_reasoning": "The user mentions Ok, thank you so much. It is not yet clear whether the user needs the full scope of th...
{ "answer": "I can help with the train. Could you share the departure, destination, day?", "reasoning": "Across the dialogue, the user is talking about ok, thank you so much. Key context: ok, thank you so much; yes, that will be fine; okay, sound correct; this can make working for me one person as well. The answer ...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1455", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004505
train
audio
multi_turn
[ { "role": "user", "text": "leave on sunday", "turn_id": 1 }, { "role": "assistant", "text": "correct.", "turn_id": 2 }, { "role": "user", "text": "yes, it's virtue to silver.", "turn_id": 3 }, { "role": "assistant", "text": "i s g e r t r u d e for georgia and...
{ "chunk_text": "Chunk 1: leave on sunday\nChunk 2: correct.\nChunk 3: yes, it's virtue to silver.\nChunk 4: i s g e r t r u d e for georgia and sylvia is s i l v e r a.\nChunk 5: uh, no, no, no, shall i give you my email address.\nChunk 6: edge. and silvera my name. and as.\nChunk 7: and s g h w.\nChunk 8: at outloo...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "leave on sunday", "role": "user", "streaming_reasoning": "The user mentions Leave on sunday. It is not yet clear whether the user needs the full scope of the user's requ...
{ "answer": "I can help with the hotel. Could you share the area, price, hotel_type?", "reasoning": "Across the dialogue, the user is talking about leave on sunday correct. Key context: leave on sunday correct; yes, it's virtue to silver; i s g e r t r u d e for georgia and sylvia is s i l v e r a; uh, no, no, no, ...
{ "category": "work_productivity", "subcategory": "meeting_summary", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0165", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004506
train
audio
multi_turn
[ { "role": "user", "text": "hello customer service.", "turn_id": 1 }, { "role": "assistant", "text": "i'm okay. and how are you.", "turn_id": 2 }, { "role": "user", "text": "i want to try some local restaurants.", "turn_id": 3 }, { "role": "assistant", "text": ...
{ "chunk_text": "Chunk 1: hello customer service.\nChunk 2: i'm okay. and how are you.\nChunk 3: i want to try some local restaurants.\nChunk 4: can i get a train that leaves on thursday.\nChunk 5: i'm headed for kingsland to leave my children.\nChunk 6: i would like to arrive by 1500.\nChunk 7: you no problem.", "...
{ "num_checkpoints": 7, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello customer service.", "role": "user", "streaming_reasoning": "The user mentions Hello customer service. It is not yet clear whether the user needs the full scope of ...
{ "answer": "I can help find a matching train option with those details.", "reasoning": "Across the dialogue, the user is reporting a problem with to try some local restaurants. Key context: hello customer service; and how are you; i want to try some local restaurants; can i get a train that leaves on thursday. The...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2840", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004507
train
audio
multi_turn
[ { "role": "user", "text": "okay.", "turn_id": 1 }, { "role": "assistant", "text": "alright, and what is the interest for you.", "turn_id": 2 }, { "role": "user", "text": "okay, fantastic. um, i do also need to book a taxi, uh, to go between guesthouse and the museum.", "t...
{ "chunk_text": "Chunk 1: okay.\nChunk 2: alright, and what is the interest for you.\nChunk 3: okay, fantastic. um, i do also need to book a taxi, uh, to go between guesthouse and the museum.\nChunk 4: yeah, i'm going to need to leave.\nChunk 5: let me give it to you. so my phone number is 6. 6, 4.\nChunk 6: 6 0 6 6....
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay.", "role": "user", "streaming_reasoning": "The user mentions Okay. It is not yet clear whether the user needs the full scope of the user's request. The assistant sh...
{ "answer": "I can help book that. Could you provide the departure, destination?", "reasoning": "Across the dialogue, the user is wanting to book taxi. Key context: alright, and what is the interest for you; okay, fantastic; um, i do also need to book a taxi, uh, to go between guesthouse and the museum; yeah, i'm g...
{ "category": "information_extraction", "subcategory": "pros_cons_extraction", "difficulty": "medium", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2929", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004508
train
audio
multi_turn
[ { "role": "user", "text": "okay. the attraction beat.", "turn_id": 1 }, { "role": "assistant", "text": "may i please have the phone number.", "turn_id": 2 }, { "role": "user", "text": "may you please provide me at the open hours.", "turn_id": 3 }, { "role": "assis...
{ "chunk_text": "Chunk 1: okay. the attraction beat.\nChunk 2: may i please have the phone number.\nChunk 3: may you please provide me at the open hours.\nChunk 4: okay, i am also looking for a place to stay.\nChunk 5: the hotel should be in the expensive price range and should include free parking. the hotel should ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "okay. the attraction beat.", "role": "user", "streaming_reasoning": "The user mentions Okay. the attraction beat. It is not yet clear whether the user needs the full sco...
{ "answer": "Sure. I can provide the area, type for the matching hotel option.", "reasoning": "Across the dialogue, the user is searching for place to stay. Key context: the attraction beat; may i please have the phone number; may you please provide me at the open hours; okay, i am also looking for a place to stay....
{ "category": "daily_life", "subcategory": "shopping_decision", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL1570", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004509
train
audio
multi_turn
[ { "role": "user", "text": "hello.", "turn_id": 1 }, { "role": "assistant", "text": "yes, i am plumbing a trip to cambridge and i need some information, please.", "turn_id": 2 }, { "role": "user", "text": "so 1st of all, i need some place to stay in um our workflow is and it n...
{ "chunk_text": "Chunk 1: hello.\nChunk 2: yes, i am plumbing a trip to cambridge and i need some information, please.\nChunk 3: so 1st of all, i need some place to stay in um our workflow is and it needs to be cheap, please.\nChunk 4: yes, and i'd like it to have four soft and i also need free wifi.", "dialogue_te...
{ "num_checkpoints": 4, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "hello.", "role": "user", "streaming_reasoning": "The user mentions Hello. It is not yet clear whether the user needs the full scope of the user's request. The assistant ...
{ "answer": "I can help with the hotel. Could you share the area, hotel_type?", "reasoning": "Across the dialogue, the user is talking about information. Key context: yes, i am plumbing a trip to cambridge and i need some information, please; so 1st of all, i need some place to stay in um our workflow is and it nee...
{ "category": "tech_support", "subcategory": "device_troubleshooting", "difficulty": "medium", "intent_type": "task_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3535", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004510
train
audio
multi_turn
[ { "role": "user", "text": "yes, i wanted somewhere convenient, so please make it in the century.", "turn_id": 1 }, { "role": "assistant", "text": "i think i just want to you know make sure i could know about the entrance fee attraction type and phone number i don't really", "turn_id": 2 ...
{ "chunk_text": "Chunk 1: yes, i wanted somewhere convenient, so please make it in the century.\nChunk 2: i think i just want to you know make sure i could know about the entrance fee attraction type and phone number i don't really\nChunk 3: care about which type\nChunk 4: okay.\nChunk 5: oh, great.\nChunk 6: um exce...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes, i wanted somewhere convenient, so please make it in the century.", "role": "user", "streaming_reasoning": "The user mentions Yes, i wanted somewhere convenient, so ...
{ "answer": "I can help with the restaurant. Could you share the area, food, price?", "reasoning": "Across the dialogue, the user is searching for restaurant. Key context: yes, i wanted somewhere convenient, so please make it in the century; i think i just want to you know make sure i could know about the entrance ...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL0661", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004511
train
audio
multi_turn
[ { "role": "user", "text": "i'm not looking for a place to stay that is in modern price range.", "turn_id": 1 }, { "role": "assistant", "text": "yes, for the location i wanted to be in the opposite of.", "turn_id": 2 }, { "role": "user", "text": "and i also want to have free p...
{ "chunk_text": "Chunk 1: i'm not looking for a place to stay that is in modern price range.\nChunk 2: yes, for the location i wanted to be in the opposite of.\nChunk 3: and i also want to have free parking.\nChunk 4: i prefer a congested.\nChunk 5: um i want to double confirm the hotel type of this thing.\nChunk 6: ...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "i'm not looking for a place to stay that is in modern price range.", "role": "user", "streaming_reasoning": "The user mentions I'm not looking for a place to stay that i...
{ "answer": "Sure. I can provide the type, stars, phone for the matching hotel option.", "reasoning": "Across the dialogue, the user is searching for place to stay that is in modern price range. Key context: i'm not looking for a place to stay that is in modern price range; yes, for the location i wanted to be in t...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "search_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "SNG0515", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004512
train
audio
multi_turn
[ { "role": "user", "text": "and harold lee.", "turn_id": 1 }, { "role": "assistant", "text": "is ha rr for ruling.", "turn_id": 2 }, { "role": "user", "text": "ol dd for date.", "turn_id": 3 }, { "role": "assistant", "text": "l y y for, yes, that's all.", "...
{ "chunk_text": "Chunk 1: and harold lee.\nChunk 2: is ha rr for ruling.\nChunk 3: ol dd for date.\nChunk 4: l y y for, yes, that's all.\nChunk 5: okay. thank you.\nChunk 6: sorry, but i don't want.\nChunk 7: yes, i also want to look for trim these.\nChunk 8: uh, i will depart from a place called, uh, it start with c...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "and harold lee.", "role": "user", "streaming_reasoning": "The user mentions And harold lee. It is not yet clear whether the user needs the full scope of the user's reque...
{ "answer": "I can help with the train. Could you share the destination, day, time?", "reasoning": "Across the dialogue, the user is talking about to look for trim these. Key context: and harold lee; is ha rr for ruling; ol dd for date; l y y for, yes, that's all. The answer should focus on the latest user turn whi...
{ "category": "daily_life", "subcategory": "home_task", "difficulty": "medium", "intent_type": "conversational_understanding" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL3934", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_spokenwoz_v02_004513
train
audio
multi_turn
[ { "role": "user", "text": "yes. um, could you please make a booking for me and two other people.", "turn_id": 1 }, { "role": "assistant", "text": "yes, it should be for me and two other people and it should be from monday. and it should be for three nights.", "turn_id": 2 }, { "r...
{ "chunk_text": "Chunk 1: yes. um, could you please make a booking for me and two other people.\nChunk 2: yes, it should be for me and two other people and it should be from monday. and it should be for three nights.\nChunk 3: okay. um it's, um estine phillips.\nChunk 4: p h io l ips.\nChunk 5: um lisa, could you als...
{ "num_checkpoints": 8, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "yes. um, could you please make a booking for me and two other people.", "role": "user", "streaming_reasoning": "The user mentions Yes. It is not yet clear whether the us...
{ "answer": "I can help find a matching restaurant option with those details.", "reasoning": "Across the dialogue, the user is searching for restaurant. Key context: um, could you please make a booking for me and two other people; yes, it should be for me and two other people and it should be from monday; and it sh...
{ "category": "daily_life", "subcategory": "personal_preference", "difficulty": "hard", "intent_type": "booking_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "SpokenWOZ", "dialogue_id": "MUL2451", "language": "en", "license": "cc-by-nc-4.0", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }
audio_multidialog_v02_001327
train
audio
multi_turn
[ { "role": "user", "text": "Human: He's gt hurt, missed a few games, Lamar took his spot, and that's all she wrote.", "turn_id": 1 }, { "role": "assistant", "text": "Assistant: What do you think of Saints quarterback Drew Brees? He's good whether he's under pressure or not.", "turn_id": 2...
{ "chunk_text": "Chunk 1: Human: He's gt hurt, missed a few games, Lamar took his spot, and that's all she wrote.\nChunk 2: Assistant: What do you think of Saints quarterback Drew Brees? He's good whether he's under pressure or not.\nChunk 3: Human: Yeah, he's been good for a long time. They had been crushing teams u...
{ "num_checkpoints": 3, "turn_checkpoints": [ { "checkpoint_id": "C1", "confidence": "medium", "observed_text": "Human: He's gt hurt, missed a few games, Lamar took his spot, and that's all she wrote.", "role": "user", "streaming_reasoning": "The user signals they may be wrapping u...
{ "answer": "I heard they wore white because the coach lost a bet to the Eagles coach.", "reasoning": "Across the dialogue, the user is asking for help with human: he's gt hurt, missed a few games, lamar took his spot, and that's all she wrote. Key context: Human: He's gt hurt, missed a few games, Lamar took his sp...
{ "category": "social_communication", "subcategory": "message_reply", "difficulty": "medium", "intent_type": "information_request" }
{ "audio_quality": "standard", "is_high_quality": true, "natural_reasoning": true, "reasoning_quality": "high", "sft_ready": true, "taxonomy_quality": "high", "timestamp_quality": "standard", "target_grounded": true }
{ "audio_id": null, "dataset": "MultiDialog", "dialogue_id": "t_96461533-b83b-4bbe-8c56-d87f885e75cd", "language": "en", "license": "cc", "turn_ids": [] }
{ "created_by": "rule_based_pipeline_with_content_grounding", "notes": null, "release_version": "vFinal" }