Upload folder using huggingface_hub

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/Step-Audio-2-mini-lora_LibriSpeech_sub.jsonl

@@ -0,0 +1,24 @@
+{"index": 0, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0003.flac", "answer": "THERE WAS SOMETHING IN HIS AIR AND MANNER THAT BETRAYED TO THE SCOUT THE UTTER CONFUSION OF THE STATE OF HIS MIND", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>There was something in his air and manner that betrayed to the scout, the utter confusion of the state of his mind.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 1, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0012.flac", "answer": "FOUR OR FIVE OF THE LATTER ONLY LINGERED ABOUT THE DOOR OF THE PRISON OF UNCAS WARY BUT CLOSE OBSERVERS OF THE MANNER OF THEIR CAPTIVE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Four or five of the latter only lingered about the door of the prison of uncas wary, but close observers of the manner of their captive.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 2, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0026.flac", "answer": "WELL WHAT CAN'T BE DONE BY MAIN COURAGE IN WAR MUST BE DONE BY CIRCUMVENTION", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Well, what can't be done by main courage in war must be done by circumvention.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 3, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0022.flac", "answer": "THE DELAWARES ARE CHILDREN OF THE TORTOISE AND THEY OUTSTRIP THE DEER", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The delawares are children of the tortoise, and they outstrip the deer.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 4, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0002.flac", "answer": "IN OTHER WORDS WHILE HE HAD IMPLICIT FAITH IN THE ABILITY OF BALAAM'S ASS TO SPEAK HE WAS SOMEWHAT SKEPTICAL ON THE SUBJECT OF A BEAR'S SINGING AND YET HE HAD BEEN ASSURED OF THE LATTER ON THE TESTIMONY OF HIS OWN EXQUISITE ORGANS", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>In other words, while he had implicit faith in the ability of balaam's ass to speak, he was somewhat sceptical on the subject of a bear's singing, and yet he had been assured of the latter on the testimony of his own exquisite organs.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 5, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0025.flac", "answer": "SO UNCAS YOU HAD BETTER TAKE THE LEAD WHILE I WILL PUT ON THE SKIN AGAIN AND TRUST TO CUNNING FOR WANT OF SPEED", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>So uncas, you had better take the lead while i will put on the skin again and trust to cunning for want of speed.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 6, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0010.flac", "answer": "THE TASK WILL NOT BE DIFFICULT RETURNED DAVID HESITATING THOUGH I GREATLY FEAR YOUR PRESENCE WOULD RATHER INCREASE THAN MITIGATE HIS UNHAPPY FORTUNES", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The task will not be difficult returned. David hesitating, though i greatly fear your presence would rather increase than mitigate his unhappy fortunes.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 7, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0001.flac", "answer": "IN HIS RETURN TO THE CAMP HIS ACUTE AND PRACTISED INTELLECTS WERE INTENTLY ENGAGED IN DEVISING MEANS TO COUNTERACT A WATCHFULNESS AND SUSPICION ON THE PART OF HIS ENEMIES THAT HE KNEW WERE IN NO DEGREE INFERIOR TO HIS OWN", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>In his return to the camp, his acute and practised intellects were intently engaged in devising means to counteract a watchfulness and suspicion on the part of his enemies that he knew were in no degree inferior to his own.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 8, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0005.flac", "answer": "THE BEAR SHOOK HIS SHAGGY SIDES AND THEN A WELL KNOWN VOICE REPLIED", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The bear shook his shaggy sides and then a well known voice replied.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 9, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0017.flac", "answer": "THEN AS IF SATISFIED OF THEIR SAFETY THE SCOUT LEFT HIS POSITION AND SLOWLY ENTERED THE PLACE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Then as if satisfied of their safety, the scout left his position and slowly entered the place.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 10, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0015.flac", "answer": "BUT THE BEAR INSTEAD OF OBEYING MAINTAINED THE SEAT IT HAD TAKEN AND GROWLED", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>But the bear, instead of obeying maintained the seat it had taken and growled.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 11, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0019.flac", "answer": "UNCAS OCCUPIED A DISTANT CORNER IN A RECLINING ATTITUDE BEING RIGIDLY BOUND BOTH HANDS AND FEET BY STRONG AND PAINFUL WITHES", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Uncas occupied a distant corner in a reclining attitude being rigidly bound, both hands and feet by strong and painful withes.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 12, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0011.flac", "answer": "THE LODGE IN WHICH UNCAS WAS CONFINED WAS IN THE VERY CENTER OF THE VILLAGE AND IN A SITUATION PERHAPS MORE DIFFICULT THAN ANY OTHER TO APPROACH OR LEAVE WITHOUT OBSERVATION", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The lodge in which uncas was confined was in the very center of the village and in a situation, perhaps more difficult than any other to approach or leave without observation.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 13, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0004.flac", "answer": "THE INGENIOUS HAWKEYE WHO RECALLED THE HASTY MANNER IN WHICH THE OTHER HAD ABANDONED HIS POST AT THE BEDSIDE OF THE SICK WOMAN WAS NOT WITHOUT HIS SUSPICIONS CONCERNING THE SUBJECT OF SO MUCH SOLEMN DELIBERATION", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The ingenious hawkeye, who recalled the hasty manner in which the other had abandoned his post at the bedside of the sick woman was not without his suspicions concerning the subject of so much solemn deliberation.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 14, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0016.flac", "answer": "THE CUNNING MAN IS AFRAID THAT HIS BREATH WILL BLOW UPON HIS BROTHERS AND TAKE AWAY THEIR COURAGE TOO CONTINUED DAVID IMPROVING THE HINT HE RECEIVED THEY MUST STAND FURTHER OFF", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>The cunning man is afraid that his breath will blow upon his brothers and take away their courage, too continued david, improving the hint he received, they must stand further off.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 15, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0007.flac", "answer": "COME COME RETURNED HAWKEYE UNCASING HIS HONEST COUNTENANCE THE BETTER TO ASSURE THE WAVERING CONFIDENCE OF HIS COMPANION YOU MAY SEE A SKIN WHICH IF IT BE NOT AS WHITE AS ONE OF THE GENTLE ONES HAS NO TINGE OF RED TO IT THAT THE WINDS OF THE HEAVEN AND THE SUN HAVE NOT BESTOWED NOW LET US TO BUSINESS", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Come come returned hawkeye, uncasing his honest countenance, the better to assure the wavering confidence of his companion. You may see a skin, which, if it be not as white as one of the gentle ones, has no tinge of red to it, that the winds of the heaven and the sun have not bestowed. Now let us to business.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 16, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0014.flac", "answer": "THEY DREW BACK A LITTLE FROM THE ENTRANCE AND MOTIONED TO THE SUPPOSED CONJURER TO ENTER", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>They drew back a little from the entrance and motioned to the supposed conjurer to enter.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 17, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0009.flac", "answer": "I GREATLY MOURN THAT ONE SO WELL DISPOSED SHOULD DIE IN HIS IGNORANCE AND I HAVE SOUGHT A GOODLY HYMN CAN YOU LEAD ME TO HIM", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>I greatly mourn that one so well disposed should die in his ignorance, and i have sought a goodly him. Can you lead me to him?", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 18, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0029.flac", "answer": "IF YOU ARE NOT THEN KNOCKED ON THE HEAD YOUR BEING A NON COMPOSSER WILL PROTECT YOU AND YOU'LL THEN HAVE A GOOD REASON TO EXPECT TO DIE IN YOUR BED", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>If you are not then knocked on the head, your being a non compos mentis will protect you, and you'll then have a good reason to expect to die in your bed.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 19, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0035.flac", "answer": "THEN HEAVING A HEAVY SIGH PROBABLY AMONG THE LAST HE EVER DREW IN PINING FOR A CONDITION HE HAD SO LONG ABANDONED HE ADDED IT IS WHAT I WOULD WISH TO PRACTISE MYSELF AS ONE WITHOUT A CROSS OF BLOOD THOUGH IT IS NOT ALWAYS EASY TO DEAL WITH AN INDIAN AS YOU WOULD WITH A FELLOW CHRISTIAN", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Then heaving a heavy sigh, probably among the last he ever drew in pining for a condition he had. So long abandoned, he added, it is what i would wish to practise myself as one without a cross of blood, though it is not always easy to deal with an indian as you would with a fellow - christian.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 20, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0023.flac", "answer": "UNCAS WHO HAD ALREADY APPROACHED THE DOOR IN READINESS TO LEAD THE WAY NOW RECOILED AND PLACED HIMSELF ONCE MORE IN THE BOTTOM OF THE LODGE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Uncas, who had already approached the door in readiness to lead the way now recoiled and placed himself once more in the bottom of the lodge.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 21, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0030.flac", "answer": "SO CHOOSE FOR YOURSELF TO MAKE A RUSH OR TARRY HERE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>So choose for yourself to make a rush or tarry here.", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 22, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0031.flac", "answer": "BRAVELY AND GENEROUSLY HAS HE BATTLED IN MY BEHALF AND THIS AND MORE WILL I DARE IN HIS SERVICE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Bravely and generously has he battled in my behalf, and this and more will i dare in his service?", "real_prompt": "请记录下你所听到的语音内容。"}
+{"index": 23, "question": "Please transcribe the spoken content into written text.", "audio_path": "/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0032.flac", "answer": "KEEP SILENT AS LONG AS MAY BE AND IT WOULD BE WISE WHEN YOU DO SPEAK TO BREAK OUT SUDDENLY IN ONE OF YOUR SHOUTINGS WHICH WILL SERVE TO REMIND THE INDIANS THAT YOU ARE NOT ALTOGETHER AS RESPONSIBLE AS MEN SHOULD BE", "subset": "test_clean", "task_type": "understanding", "prediction": "<英语>Keep silent as long as may be, and it would be wise when you do speak to break out suddenly in one of your shoutings, which will serve to remind the indians that you are not altogether as responsible as men should be.", "real_prompt": "请记录下你所听到的语音内容。"}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/Step-Audio-2-mini-lora_LibriSpeech_sub_default_performance.json

@@ -0,0 +1,13 @@
+{
+    "task": "ASR",
+    "dataset": "LibriSpeech_sub",
+    "model": "Step-Audio-2-mini-lora",
+    "date": "2025-12-25 05:41:25.173571",
+    "performance": {
+        "test_clean": {
+            "wer": 0.45,
+            "total": 24
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/Step-Audio-2-mini-lora_LibriSpeech_sub_wer_details.jsonl

@@ -0,0 +1,24 @@
+{"index":0,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0003.flac","answer":"THERE WAS SOMETHING IN HIS AIR AND MANNER THAT BETRAYED TO THE SCOUT THE UTTER CONFUSION OF THE STATE OF HIS MIND","subset":"test_clean","task_type":"understanding","prediction":"<英语>There was something in his air and manner that betrayed to the scout, the utter confusion of the state of his mind.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":22,"utt_wer":0.0},"utt_wer":0.0}
+{"index":1,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0012.flac","answer":"FOUR OR FIVE OF THE LATTER ONLY LINGERED ABOUT THE DOOR OF THE PRISON OF UNCAS WARY BUT CLOSE OBSERVERS OF THE MANNER OF THEIR CAPTIVE","subset":"test_clean","task_type":"understanding","prediction":"<英语>Four or five of the latter only lingered about the door of the prison of uncas wary, but close observers of the manner of their captive.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":26,"utt_wer":0.0},"utt_wer":0.0}
+{"index":2,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0026.flac","answer":"WELL WHAT CAN'T BE DONE BY MAIN COURAGE IN WAR MUST BE DONE BY CIRCUMVENTION","subset":"test_clean","task_type":"understanding","prediction":"<英语>Well, what can't be done by main courage in war must be done by circumvention.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":16,"utt_wer":0.0},"utt_wer":0.0}
+{"index":3,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0022.flac","answer":"THE DELAWARES ARE CHILDREN OF THE TORTOISE AND THEY OUTSTRIP THE DEER","subset":"test_clean","task_type":"understanding","prediction":"<英语>The delawares are children of the tortoise, and they outstrip the deer.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":12,"utt_wer":0.0},"utt_wer":0.0}
+{"index":4,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0002.flac","answer":"IN OTHER WORDS WHILE HE HAD IMPLICIT FAITH IN THE ABILITY OF BALAAM'S ASS TO SPEAK HE WAS SOMEWHAT SKEPTICAL ON THE SUBJECT OF A BEAR'S SINGING AND YET HE HAD BEEN ASSURED OF THE LATTER ON THE TESTIMONY OF HIS OWN EXQUISITE ORGANS","subset":"test_clean","task_type":"understanding","prediction":"<英语>In other words, while he had implicit faith in the ability of balaam's ass to speak, he was somewhat sceptical on the subject of a bear's singing, and yet he had been assured of the latter on the testimony of his own exquisite organs.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":46,"utt_wer":0.0},"utt_wer":0.0}
+{"index":5,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0025.flac","answer":"SO UNCAS YOU HAD BETTER TAKE THE LEAD WHILE I WILL PUT ON THE SKIN AGAIN AND TRUST TO CUNNING FOR WANT OF SPEED","subset":"test_clean","task_type":"understanding","prediction":"<英语>So uncas, you had better take the lead while i will put on the skin again and trust to cunning for want of speed.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":24,"utt_wer":0.0},"utt_wer":0.0}
+{"index":6,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0010.flac","answer":"THE TASK WILL NOT BE DIFFICULT RETURNED DAVID HESITATING THOUGH I GREATLY FEAR YOUR PRESENCE WOULD RATHER INCREASE THAN MITIGATE HIS UNHAPPY FORTUNES","subset":"test_clean","task_type":"understanding","prediction":"<英语>The task will not be difficult returned. David hesitating, though i greatly fear your presence would rather increase than mitigate his unhappy fortunes.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":23,"utt_wer":0.0},"utt_wer":0.0}
+{"index":7,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0001.flac","answer":"IN HIS RETURN TO THE CAMP HIS ACUTE AND PRACTISED INTELLECTS WERE INTENTLY ENGAGED IN DEVISING MEANS TO COUNTERACT A WATCHFULNESS AND SUSPICION ON THE PART OF HIS ENEMIES THAT HE KNEW WERE IN NO DEGREE INFERIOR TO HIS OWN","subset":"test_clean","task_type":"understanding","prediction":"<英语>In his return to the camp, his acute and practised intellects were intently engaged in devising means to counteract a watchfulness and suspicion on the part of his enemies that he knew were in no degree inferior to his own.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":40,"utt_wer":0.0},"utt_wer":0.0}
+{"index":8,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0005.flac","answer":"THE BEAR SHOOK HIS SHAGGY SIDES AND THEN A WELL KNOWN VOICE REPLIED","subset":"test_clean","task_type":"understanding","prediction":"<英语>The bear shook his shaggy sides and then a well known voice replied.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":13,"utt_wer":0.0},"utt_wer":0.0}
+{"index":9,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0017.flac","answer":"THEN AS IF SATISFIED OF THEIR SAFETY THE SCOUT LEFT HIS POSITION AND SLOWLY ENTERED THE PLACE","subset":"test_clean","task_type":"understanding","prediction":"<英语>Then as if satisfied of their safety, the scout left his position and slowly entered the place.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":17,"utt_wer":0.0},"utt_wer":0.0}
+{"index":10,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0015.flac","answer":"BUT THE BEAR INSTEAD OF OBEYING MAINTAINED THE SEAT IT HAD TAKEN AND GROWLED","subset":"test_clean","task_type":"understanding","prediction":"<英语>But the bear, instead of obeying maintained the seat it had taken and growled.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":14,"utt_wer":0.0},"utt_wer":0.0}
+{"index":11,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0019.flac","answer":"UNCAS OCCUPIED A DISTANT CORNER IN A RECLINING ATTITUDE BEING RIGIDLY BOUND BOTH HANDS AND FEET BY STRONG AND PAINFUL WITHES","subset":"test_clean","task_type":"understanding","prediction":"<英语>Uncas occupied a distant corner in a reclining attitude being rigidly bound, both hands and feet by strong and painful withes.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":21,"utt_wer":0.0},"utt_wer":0.0}
+{"index":12,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0011.flac","answer":"THE LODGE IN WHICH UNCAS WAS CONFINED WAS IN THE VERY CENTER OF THE VILLAGE AND IN A SITUATION PERHAPS MORE DIFFICULT THAN ANY OTHER TO APPROACH OR LEAVE WITHOUT OBSERVATION","subset":"test_clean","task_type":"understanding","prediction":"<英语>The lodge in which uncas was confined was in the very center of the village and in a situation, perhaps more difficult than any other to approach or leave without observation.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":31,"utt_wer":0.0},"utt_wer":0.0}
+{"index":13,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0004.flac","answer":"THE INGENIOUS HAWKEYE WHO RECALLED THE HASTY MANNER IN WHICH THE OTHER HAD ABANDONED HIS POST AT THE BEDSIDE OF THE SICK WOMAN WAS NOT WITHOUT HIS SUSPICIONS CONCERNING THE SUBJECT OF SO MUCH SOLEMN DELIBERATION","subset":"test_clean","task_type":"understanding","prediction":"<英语>The ingenious hawkeye, who recalled the hasty manner in which the other had abandoned his post at the bedside of the sick woman was not without his suspicions concerning the subject of so much solemn deliberation.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":36,"utt_wer":0.0},"utt_wer":0.0}
+{"index":14,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0016.flac","answer":"THE CUNNING MAN IS AFRAID THAT HIS BREATH WILL BLOW UPON HIS BROTHERS AND TAKE AWAY THEIR COURAGE TOO CONTINUED DAVID IMPROVING THE HINT HE RECEIVED THEY MUST STAND FURTHER OFF","subset":"test_clean","task_type":"understanding","prediction":"<英语>The cunning man is afraid that his breath will blow upon his brothers and take away their courage, too continued david, improving the hint he received, they must stand further off.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":31,"utt_wer":0.0},"utt_wer":0.0}
+{"index":15,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0007.flac","answer":"COME COME RETURNED HAWKEYE UNCASING HIS HONEST COUNTENANCE THE BETTER TO ASSURE THE WAVERING CONFIDENCE OF HIS COMPANION YOU MAY SEE A SKIN WHICH IF IT BE NOT AS WHITE AS ONE OF THE GENTLE ONES HAS NO TINGE OF RED TO IT THAT THE WINDS OF THE HEAVEN AND THE SUN HAVE NOT BESTOWED NOW LET US TO BUSINESS","subset":"test_clean","task_type":"understanding","prediction":"<英语>Come come returned hawkeye, uncasing his honest countenance, the better to assure the wavering confidence of his companion. You may see a skin, which, if it be not as white as one of the gentle ones, has no tinge of red to it, that the winds of the heaven and the sun have not bestowed. Now let us to business.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":60,"utt_wer":0.0},"utt_wer":0.0}
+{"index":16,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0014.flac","answer":"THEY DREW BACK A LITTLE FROM THE ENTRANCE AND MOTIONED TO THE SUPPOSED CONJURER TO ENTER","subset":"test_clean","task_type":"understanding","prediction":"<英语>They drew back a little from the entrance and motioned to the supposed conjurer to enter.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":16,"utt_wer":0.0},"utt_wer":0.0}
+{"index":17,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0009.flac","answer":"I GREATLY MOURN THAT ONE SO WELL DISPOSED SHOULD DIE IN HIS IGNORANCE AND I HAVE SOUGHT A GOODLY HYMN CAN YOU LEAD ME TO HIM","subset":"test_clean","task_type":"understanding","prediction":"<英语>I greatly mourn that one so well disposed should die in his ignorance, and i have sought a goodly him. Can you lead me to him?","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":1,"nref":26,"utt_wer":0.0384615385},"utt_wer":3.85}
+{"index":18,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0029.flac","answer":"IF YOU ARE NOT THEN KNOCKED ON THE HEAD YOUR BEING A NON COMPOSSER WILL PROTECT YOU AND YOU'LL THEN HAVE A GOOD REASON TO EXPECT TO DIE IN YOUR BED","subset":"test_clean","task_type":"understanding","prediction":"<英语>If you are not then knocked on the head, your being a non compos mentis will protect you, and you'll then have a good reason to expect to die in your bed.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":2,"nref":32,"utt_wer":0.0625},"utt_wer":6.25}
+{"index":19,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0035.flac","answer":"THEN HEAVING A HEAVY SIGH PROBABLY AMONG THE LAST HE EVER DREW IN PINING FOR A CONDITION HE HAD SO LONG ABANDONED HE ADDED IT IS WHAT I WOULD WISH TO PRACTISE MYSELF AS ONE WITHOUT A CROSS OF BLOOD THOUGH IT IS NOT ALWAYS EASY TO DEAL WITH AN INDIAN AS YOU WOULD WITH A FELLOW CHRISTIAN","subset":"test_clean","task_type":"understanding","prediction":"<英语>Then heaving a heavy sigh, probably among the last he ever drew in pining for a condition he had. So long abandoned, he added, it is what i would wish to practise myself as one without a cross of blood, though it is not always easy to deal with an indian as you would with a fellow - christian.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":58,"utt_wer":0.0},"utt_wer":0.0}
+{"index":20,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0023.flac","answer":"UNCAS WHO HAD ALREADY APPROACHED THE DOOR IN READINESS TO LEAD THE WAY NOW RECOILED AND PLACED HIMSELF ONCE MORE IN THE BOTTOM OF THE LODGE","subset":"test_clean","task_type":"understanding","prediction":"<英语>Uncas, who had already approached the door in readiness to lead the way now recoiled and placed himself once more in the bottom of the lodge.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":26,"utt_wer":0.0},"utt_wer":0.0}
+{"index":21,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0030.flac","answer":"SO CHOOSE FOR YOURSELF TO MAKE A RUSH OR TARRY HERE","subset":"test_clean","task_type":"understanding","prediction":"<英语>So choose for yourself to make a rush or tarry here.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":11,"utt_wer":0.0},"utt_wer":0.0}
+{"index":22,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0031.flac","answer":"BRAVELY AND GENEROUSLY HAS HE BATTLED IN MY BEHALF AND THIS AND MORE WILL I DARE IN HIS SERVICE","subset":"test_clean","task_type":"understanding","prediction":"<英语>Bravely and generously has he battled in my behalf, and this and more will i dare in his service?","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":19,"utt_wer":0.0},"utt_wer":0.0}
+{"index":23,"question":"Please transcribe the spoken content into written text.","audio_path":"\/workspace\/intern\/pangkaiyu\/Kimi-Audio\/dataset\/LibriSpeech\/librispeech\/LibriSpeech\/test-clean\/1320\/122617\/1320-122617-0032.flac","answer":"KEEP SILENT AS LONG AS MAY BE AND IT WOULD BE WISE WHEN YOU DO SPEAK TO BREAK OUT SUDDENLY IN ONE OF YOUR SHOUTINGS WHICH WILL SERVE TO REMIND THE INDIANS THAT YOU ARE NOT ALTOGETHER AS RESPONSIBLE AS MEN SHOULD BE","subset":"test_clean","task_type":"understanding","prediction":"<英语>Keep silent as long as may be, and it would be wise when you do speak to break out suddenly in one of your shoutings, which will serve to remind the indians that you are not altogether as responsible as men should be.","real_prompt":"请记录下你所听到的语音内容。","wer_details":{"err":0,"nref":43,"utt_wer":0.0},"utt_wer":0.0}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank0.log

@@ -0,0 +1,7 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 0, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0003.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}
+2025-12-25 05:41:14 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-25 05:41:24 | INFO | waiting for other ranks to finish, time elapsed: 20s
+2025-12-25 05:41:25 | INFO | model Step-Audio-2-mini-lora, data LibriSpeech_sub, all 8 result merged to lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/Step-Audio-2-mini-lora_LibriSpeech_sub.jsonl.
+2025-12-25 05:41:25 | INFO | evaluating result of LibriSpeech_sub with method: qwen2-audio-impl
+2025-12-25 05:41:25 | INFO | model Step-Audio-2-mini-lora, data LibriSpeech_sub evaluated.

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0012.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0026.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0022.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0002.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0025.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0010.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/LibriSpeech_sub/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-25 05:40:52 | INFO | Running Step-Audio-2-mini-lora on dataset: LibriSpeech_sub
+2025-12-25 05:40:52 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/dataset/LibriSpeech/librispeech/LibriSpeech/test-clean/1320/122617/1320-122617-0001.flac'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'LibriSpeech', 'dataset_name': 'LibriSpeech_sub', 'lang': 'en', 'subset': 'test_clean'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/Step-Audio-2-mini-lora_chime4_dev-real_kimi_default_performance.json

@@ -0,0 +1,25 @@
+{
+    "task": "ASR",
+    "dataset": "chime4_dev-real_kimi",
+    "model": "Step-Audio-2-mini-lora",
+    "date": "2025-12-25 06:02:13.050791",
+    "performance": {
+        "bus": {
+            "wer": 4.14,
+            "total": 410
+        },
+        "caf": {
+            "wer": 3.87,
+            "total": 410
+        },
+        "ped": {
+            "wer": 3.47,
+            "total": 410
+        },
+        "str": {
+            "wer": 3.89,
+            "total": 410
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank0.log

@@ -0,0 +1,10 @@
+2025-12-25 05:46:56 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:56 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C0103_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}
+2025-12-25 05:50:55 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-25 05:51:05 | INFO | waiting for other ranks to finish, time elapsed: 20s
+2025-12-25 05:51:15 | INFO | waiting for other ranks to finish, time elapsed: 30s
+2025-12-25 05:51:25 | INFO | waiting for other ranks to finish, time elapsed: 40s
+2025-12-25 05:51:35 | INFO | waiting for other ranks to finish, time elapsed: 50s
+2025-12-25 05:51:45 | INFO | waiting for other ranks to finish, time elapsed: 60s
+2025-12-25 05:51:45 | INFO | model Step-Audio-2-mini-lora, data chime4_dev-real_kimi, all 8 result merged to lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/Step-Audio-2-mini-lora_chime4_dev-real_kimi.jsonl.
+2025-12-25 05:51:45 | INFO | skip eval for chime4_dev-real_kimi

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:43 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:43 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C0105_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:50 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:50 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010C_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:48 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:48 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010G_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:48 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:48 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010J_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:47 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:47 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010K_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:49 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:49 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010L_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_dev-real_kimi/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-25 05:46:47 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_dev-real_kimi
+2025-12-25 05:46:47 | INFO | Msg example: {'index': 8, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/dt05_bus_real/F01_050C010O_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-dev-real', 'dataset_name': 'chime4_dev-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/Step-Audio-2-mini-lora_chime4_test-real_kimi_default_performance.json

@@ -0,0 +1,25 @@
+{
+    "task": "ASR",
+    "dataset": "chime4_test-real_kimi",
+    "model": "Step-Audio-2-mini-lora",
+    "date": "2025-12-25 06:02:13.816145",
+    "performance": {
+        "bus": {
+            "wer": 6.9,
+            "total": 330
+        },
+        "caf": {
+            "wer": 5.22,
+            "total": 330
+        },
+        "ped": {
+            "wer": 5.14,
+            "total": 330
+        },
+        "str": {
+            "wer": 4.36,
+            "total": 330
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank0.log

@@ -0,0 +1,11 @@
+2025-12-25 05:57:24 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:24 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C0202_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}
+2025-12-25 06:00:39 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-25 06:00:49 | INFO | waiting for other ranks to finish, time elapsed: 20s
+2025-12-25 06:00:59 | INFO | waiting for other ranks to finish, time elapsed: 30s
+2025-12-25 06:01:09 | INFO | waiting for other ranks to finish, time elapsed: 40s
+2025-12-25 06:01:19 | INFO | waiting for other ranks to finish, time elapsed: 50s
+2025-12-25 06:01:29 | INFO | waiting for other ranks to finish, time elapsed: 60s
+2025-12-25 06:01:39 | INFO | waiting for other ranks to finish, time elapsed: 70s
+2025-12-25 06:01:40 | INFO | model Step-Audio-2-mini-lora, data chime4_test-real_kimi, all 8 result merged to lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/Step-Audio-2-mini-lora_chime4_test-real_kimi.jsonl.
+2025-12-25 06:01:40 | INFO | skip eval for chime4_test-real_kimi

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:22 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:22 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C0204_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:24 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:24 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C0209_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:24 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:24 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C020E_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:25 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:25 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C020F_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:25 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:25 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C020G_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:24 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:24 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C020H_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/chime4_test-real_kimi/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-25 05:57:24 | INFO | Running Step-Audio-2-mini-lora on dataset: chime4_test-real_kimi
+2025-12-25 05:57:24 | INFO | Msg example: {'index': 8, 'audio': ['/workspace/intern/pangkaiyu/dg/CHIME-4/CHiME4/data/audio/16kHz/isolated_1ch_track/et05_bus_real/F05_440C020I_BUS.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'chime4-test-real', 'dataset_name': 'chime4_test-real_kimi', 'lang': 'en', 'subset': 'bus'}}

lora_test/Step-Audio-2-mini-lora/noizeus/Step-Audio-2-mini-lora_noizeus_default_performance.json

@@ -0,0 +1,121 @@
+{
+    "task": "ASR",
+    "dataset": "noizeus",
+    "model": "Step-Audio-2-mini-lora",
+    "date": "2025-12-25 06:02:12.074020",
+    "performance": {
+        "airport_0dB": {
+            "wer": 25.21,
+            "total": 30
+        },
+        "airport_10dB": {
+            "wer": 1.24,
+            "total": 30
+        },
+        "airport_15dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "airport_5dB": {
+            "wer": 10.33,
+            "total": 30
+        },
+        "babble_0dB": {
+            "wer": 39.67,
+            "total": 30
+        },
+        "babble_10dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "babble_15dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "babble_5dB": {
+            "wer": 5.37,
+            "total": 30
+        },
+        "car_0dB": {
+            "wer": 35.12,
+            "total": 30
+        },
+        "car_10dB": {
+            "wer": 6.61,
+            "total": 30
+        },
+        "car_15dB": {
+            "wer": 2.89,
+            "total": 30
+        },
+        "car_5dB": {
+            "wer": 7.02,
+            "total": 30
+        },
+        "exhibition_0dB": {
+            "wer": 28.1,
+            "total": 30
+        },
+        "exhibition_10dB": {
+            "wer": 3.31,
+            "total": 30
+        },
+        "exhibition_15dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "exhibition_5dB": {
+            "wer": 7.44,
+            "total": 30
+        },
+        "restaurant_0dB": {
+            "wer": 32.23,
+            "total": 30
+        },
+        "restaurant_10dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "restaurant_15dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "restaurant_5dB": {
+            "wer": 7.02,
+            "total": 30
+        },
+        "station_0dB": {
+            "wer": 22.73,
+            "total": 30
+        },
+        "station_10dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "station_15dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "station_5dB": {
+            "wer": 11.98,
+            "total": 30
+        },
+        "street_0dB": {
+            "wer": 35.95,
+            "total": 30
+        },
+        "street_10dB": {
+            "wer": 4.96,
+            "total": 30
+        },
+        "street_15dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "street_5dB": {
+            "wer": 14.05,
+            "total": 30
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank0.log

@@ -0,0 +1,6 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 0, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp01_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}
+2025-12-25 05:46:46 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-25 05:46:56 | INFO | waiting for other ranks to finish, time elapsed: 20s
+2025-12-25 05:46:56 | INFO | model Step-Audio-2-mini-lora, data noizeus, all 8 result merged to lora_test/Step-Audio-2-mini-lora/noizeus/Step-Audio-2-mini-lora_noizeus.jsonl.
+2025-12-25 05:46:56 | INFO | skip eval for noizeus

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:20 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:20 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp02_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp03_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp04_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp05_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp06_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp07_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}

lora_test/Step-Audio-2-mini-lora/noizeus/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-25 05:45:21 | INFO | Running Step-Audio-2-mini-lora on dataset: noizeus
+2025-12-25 05:45:21 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/Kimi-Audio/Kimi-Audio-Evalkit/data/downloaded_datasets/noizeus/noizeus/airport/0dB/sp08_airport_sn0.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'noizeus', 'dataset_name': 'noizeus', 'lang': 'en', 'subset': 'airport_0dB'}}