| [ | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_002.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips feels like it's in the back zone? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_092.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips is more likely to be judged as frontal? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_067.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one sounds like it's in front, based on loudness and clarity? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_077.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip sounds like it's in front, even with minimal cues? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_072.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip gives the sensation of being in front? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_019.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound seems to be in front, not diffused or ambient? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_001.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which segment seems to be located in the front of the stereo field? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_039.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Listen carefully: which clip appears to be located in back? There’s a 0.5-second pause between them. Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_079.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound seems to be in front, not attenuated for rear playback? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_050.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one sounds like it's in front, not processed for rear channels? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_041.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Two clips are joined by 0.5 seconds of silence. Which one sounds back to you? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_038.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Compare the spatial position of two clips with 0.5 seconds of silence. Which one sounds more forward? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_037.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips feels like it's in back, not distant? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_083.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one seems to be in back, based on auditory localization? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_044.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound appears to be in front, based on spectral cues? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_060.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip gives the impression of being in front? The two are separated by 0.5 seconds of silence. Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_090.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one has a timbral or spatial quality suggesting it's in back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_016.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound is more likely to be perceived as originating from the back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_071.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one appears to be in front, not from the rear hemisphere? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_005.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Listen to the pair: which one seems to be coming from the back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_012.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one is perceived as being in back, not behind? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_023.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip sounds like it's not behind but in back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_052.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound feels like it's right in front of you? The clips are separated by 0.5 seconds. Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_089.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one gives the sensation of being in front, not behind the listener? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_084.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound is more likely to be localized in the front? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_056.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip seems to be originating from a closer, more frontal location? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_029.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one seems to be in front, not delayed for surround effect? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_022.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound is more likely to be localized as frontal in binaural hearing? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_020.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound seems to be leading from the back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_064.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "From auditory cues, which clip seems to come from a more forward position? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_010.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Based on spatial perception, which of the two clips sounds more backward? They are separated by 0.5 seconds of silence. Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_076.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip sounds like it's in front, not from the sides or back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_027.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips seems to be presented from a front-facing source? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_009.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips feels like it's in the front zone? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_003.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound seems to be in front, not part of the ambient background? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_085.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips is more likely to be heard as frontal in headphones? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_033.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one sounds like it's in back, based on precedence effect? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_099.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound is more likely to be heard as frontal in a virtual space? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_040.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip appears to be in front, based on direct-to-reverberant ratio? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_047.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip sounds like it's in back, not arriving from behind? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_045.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip seems to be in back, not from behind the median plane? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_036.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips sounds nearer in depth? They are separated by 0.5 seconds of silence. Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_035.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip appears to be in back, not downmixed or redirected? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_057.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which of the two clips seems to be in front, not delayed or phase-shifted? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_080.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which one seems to be in back, even with identical content? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_007.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "You will hear two sounds with a half-second gap. Which one seems to come from a more frontal position? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_066.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which clip gives the impression of being in back, not virtualized for back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_087.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound gives the strongest cue for frontal localization? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_028.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound feels like it's projecting from the back? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "B" | |
| } | |
| ] | |
| }, | |
| { | |
| "voice": [ | |
| "direction/task_comparison/direction_pair_063.wav" | |
| ], | |
| "conversations": [ | |
| { | |
| "from": "human", | |
| "value": "Which sound seems to be ahead in the spatial layout? Only answer letter 'A' (refers to the first clip) or 'B' (refers to the second clip). Do not add any explanation, punctuation, or extra text. <audio>" | |
| }, | |
| { | |
| "from": "gpt", | |
| "value": "A" | |
| } | |
| ] | |
| } | |
| ] |