Add files using upload-large-folder tool

LICENSE

@@ -0,0 +1,8 @@
+The MIT License (MIT)
+Copyright © 2025 Moonshot AI
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora2/voices_dev_test/logs/rank5.log

@@ -0,0 +1,2 @@
+2026-01-04 11:37:52 | INFO | Running Step-Audio-2-mini-all-lora2 on dataset: voices_dev_test
+2026-01-04 11:37:52 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp1447/Lab41-SRI-VOiCES-rm4-babb-sp1447-ch130550-sg0026-mc05-stu-far-dg010.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'voices_dev', 'dataset_name': 'voices_dev_test', 'lang': 'en', 'subset': 'rm4-babb-far'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora2/voices_dev_test/logs/rank6.log

@@ -0,0 +1,2 @@
+2026-01-04 11:38:05 | INFO | Running Step-Audio-2-mini-all-lora2 on dataset: voices_dev_test
+2026-01-04 11:38:05 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp1447/Lab41-SRI-VOiCES-rm4-babb-sp1447-ch130551-sg0027-mc01-stu-clo-dg140.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'voices_dev', 'dataset_name': 'voices_dev_test', 'lang': 'en', 'subset': 'rm4-babb-clo'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/Step-Audio-2-mini-all-lora3_chime4_dev-real_kimi_default_performance.json

@@ -0,0 +1,25 @@
+{
+    "task": "ASR",
+    "dataset": "chime4_dev-real_kimi",
+    "model": "Step-Audio-2-mini-all-lora3",
+    "date": "2026-01-04 15:40:14.053795",
+    "performance": {
+        "bus": {
+            "wer": 4.41,
+            "total": 410
+        },
+        "caf": {
+            "wer": 4.05,
+            "total": 410
+        },
+        "ped": {
+            "wer": 3.76,
+            "total": 410
+        },
+        "str": {
+            "wer": 4.0,
+            "total": 410
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank0.log

@@ -0,0 +1,5 @@
+2026-01-04 12:15:41 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:41 | 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'}}
+2026-01-04 12:18:38 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2026-01-04 12:18:38 | INFO | model Step-Audio-2-mini-all-lora3, data chime4_dev-real_kimi, all 8 result merged to mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/Step-Audio-2-mini-all-lora3_chime4_dev-real_kimi.jsonl.
+2026-01-04 12:18:38 | INFO | skip eval for chime4_dev-real_kimi

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank1.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:29 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:29 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank2.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:34 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:34 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank3.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:30 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:30 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank4.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:31 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:31 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank5.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:30 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:30 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank6.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:31 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:31 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_dev-real_kimi/logs/rank7.log

@@ -0,0 +1,2 @@
+2026-01-04 12:15:29 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_dev-real_kimi
+2026-01-04 12:15:29 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/Step-Audio-2-mini-all-lora3_chime4_test-real_kimi_default_performance.json

@@ -0,0 +1,25 @@
+{
+    "task": "ASR",
+    "dataset": "chime4_test-real_kimi",
+    "model": "Step-Audio-2-mini-all-lora3",
+    "date": "2026-01-04 15:40:14.829699",
+    "performance": {
+        "bus": {
+            "wer": 7.37,
+            "total": 330
+        },
+        "caf": {
+            "wer": 5.28,
+            "total": 330
+        },
+        "ped": {
+            "wer": 5.33,
+            "total": 330
+        },
+        "str": {
+            "wer": 4.32,
+            "total": 330
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank0.log

@@ -0,0 +1,4 @@
+2026-01-04 12:18:38 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:38 | 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'}}
+2026-01-04 12:20:54 | INFO | model Step-Audio-2-mini-all-lora3, data chime4_test-real_kimi, all 8 result merged to mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/Step-Audio-2-mini-all-lora3_chime4_test-real_kimi.jsonl.
+2026-01-04 12:20:54 | INFO | skip eval for chime4_test-real_kimi

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank1.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:24 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:24 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank2.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:32 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:32 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank3.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:33 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:33 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank4.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:31 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:31 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank5.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:24 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:24 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank6.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:29 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:29 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/chime4_test-real_kimi/logs/rank7.log

@@ -0,0 +1,2 @@
+2026-01-04 12:18:27 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: chime4_test-real_kimi
+2026-01-04 12:18:27 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/Step-Audio-2-mini-all-lora3_noizeus_default_performance.json

@@ -0,0 +1,121 @@
+{
+    "task": "ASR",
+    "dataset": "noizeus",
+    "model": "Step-Audio-2-mini-all-lora3",
+    "date": "2026-01-04 15:40:13.064206",
+    "performance": {
+        "airport_0dB": {
+            "wer": 25.21,
+            "total": 30
+        },
+        "airport_10dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "airport_15dB": {
+            "wer": 1.65,
+            "total": 30
+        },
+        "airport_5dB": {
+            "wer": 9.92,
+            "total": 30
+        },
+        "babble_0dB": {
+            "wer": 38.43,
+            "total": 30
+        },
+        "babble_10dB": {
+            "wer": 3.31,
+            "total": 30
+        },
+        "babble_15dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "babble_5dB": {
+            "wer": 4.96,
+            "total": 30
+        },
+        "car_0dB": {
+            "wer": 35.54,
+            "total": 30
+        },
+        "car_10dB": {
+            "wer": 5.79,
+            "total": 30
+        },
+        "car_15dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "car_5dB": {
+            "wer": 7.85,
+            "total": 30
+        },
+        "exhibition_0dB": {
+            "wer": 31.4,
+            "total": 30
+        },
+        "exhibition_10dB": {
+            "wer": 4.55,
+            "total": 30
+        },
+        "exhibition_15dB": {
+            "wer": 2.48,
+            "total": 30
+        },
+        "exhibition_5dB": {
+            "wer": 7.44,
+            "total": 30
+        },
+        "restaurant_0dB": {
+            "wer": 30.17,
+            "total": 30
+        },
+        "restaurant_10dB": {
+            "wer": 1.65,
+            "total": 30
+        },
+        "restaurant_15dB": {
+            "wer": 2.89,
+            "total": 30
+        },
+        "restaurant_5dB": {
+            "wer": 6.61,
+            "total": 30
+        },
+        "station_0dB": {
+            "wer": 23.97,
+            "total": 30
+        },
+        "station_10dB": {
+            "wer": 3.31,
+            "total": 30
+        },
+        "station_15dB": {
+            "wer": 2.07,
+            "total": 30
+        },
+        "station_5dB": {
+            "wer": 9.92,
+            "total": 30
+        },
+        "street_0dB": {
+            "wer": 39.26,
+            "total": 30
+        },
+        "street_10dB": {
+            "wer": 5.37,
+            "total": 30
+        },
+        "street_15dB": {
+            "wer": 1.65,
+            "total": 30
+        },
+        "street_5dB": {
+            "wer": 13.64,
+            "total": 30
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank0.log

@@ -0,0 +1,5 @@
+2026-01-04 12:14:35 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:35 | 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'}}
+2026-01-04 12:15:41 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2026-01-04 12:15:41 | INFO | model Step-Audio-2-mini-all-lora3, data noizeus, all 8 result merged to mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/Step-Audio-2-mini-all-lora3_noizeus.jsonl.
+2026-01-04 12:15:41 | INFO | skip eval for noizeus

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank1.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:36 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:36 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank2.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:36 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:36 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank3.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:36 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:36 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank4.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:35 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:35 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank5.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:35 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:35 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank6.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:35 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:35 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/noizeus/logs/rank7.log

@@ -0,0 +1,2 @@
+2026-01-04 12:14:35 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: noizeus
+2026-01-04 12:14:35 | 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'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/Step-Audio-2-mini-all-lora3_tedlium_test1_default_performance.json

@@ -0,0 +1,13 @@
+{
+    "task": "ASR",
+    "dataset": "tedlium_test1",
+    "model": "Step-Audio-2-mini-all-lora3",
+    "date": "2026-01-04 15:40:15.722457",
+    "performance": {
+        "TEDLIUM-test": {
+            "wer": 2.54,
+            "total": 1155
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank0.log

@@ -0,0 +1,5 @@
+2026-01-04 12:20:54 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:54 | INFO | Msg example: {'index': 0, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-16.11-27.84-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}
+2026-01-04 12:24:04 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2026-01-04 12:24:04 | INFO | model Step-Audio-2-mini-all-lora3, data tedlium_test1, all 8 result merged to mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/Step-Audio-2-mini-all-lora3_tedlium_test1.jsonl.
+2026-01-04 12:24:04 | INFO | skip eval for tedlium_test1

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank1.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:39 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:39 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-27.84-38.316-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank2.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:53 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:53 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-38.316-47.64-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank3.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:50 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:50 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-47.64-57.27-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank4.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:49 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:49 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-57.27-63.88-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank5.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:38 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:38 | INFO | Msg example: {'index': 5, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-63.88-71.56-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank6.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:48 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:48 | INFO | Msg example: {'index': 6, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-71.56-79.704-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/tedlium_test1/logs/rank7.log

@@ -0,0 +1,2 @@
+2026-01-04 12:20:48 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: tedlium_test1
+2026-01-04 12:20:48 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/dg/tedlium_release1_data/test/wav/MichaelSpecter-79.704-87.33-<o,f0,male>.wav'], 'text': 'Please transcribe the audio content into text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'tedlium', 'dataset_name': 'tedlium_test1', 'lang': 'en', 'subset': 'TEDLIUM-test'}}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/voices_dev_test/Step-Audio-2-mini-all-lora3_voices_dev_test_default_performance.json

@@ -0,0 +1,137 @@
+{
+    "task": "ASR",
+    "dataset": "voices_dev_test",
+    "model": "Step-Audio-2-mini-all-lora3",
+    "date": "2026-01-04 15:40:23.816430",
+    "performance": {
+        "rm1-babb-clo": {
+            "wer": 1.3,
+            "total": 200
+        },
+        "rm1-babb-far": {
+            "wer": 2.79,
+            "total": 200
+        },
+        "rm1-musi-clo": {
+            "wer": 1.44,
+            "total": 200
+        },
+        "rm1-musi-far": {
+            "wer": 2.12,
+            "total": 200
+        },
+        "rm1-none-clo": {
+            "wer": 1.22,
+            "total": 200
+        },
+        "rm1-none-far": {
+            "wer": 1.72,
+            "total": 200
+        },
+        "rm1-tele-clo": {
+            "wer": 1.49,
+            "total": 200
+        },
+        "rm1-tele-far": {
+            "wer": 2.36,
+            "total": 200
+        },
+        "rm2-babb-clo": {
+            "wer": 1.5,
+            "total": 200
+        },
+        "rm2-babb-far": {
+            "wer": 2.94,
+            "total": 200
+        },
+        "rm2-musi-clo": {
+            "wer": 1.69,
+            "total": 200
+        },
+        "rm2-musi-far": {
+            "wer": 2.47,
+            "total": 200
+        },
+        "rm2-none-clo": {
+            "wer": 1.31,
+            "total": 200
+        },
+        "rm2-none-far": {
+            "wer": 1.75,
+            "total": 200
+        },
+        "rm2-tele-clo": {
+            "wer": 1.58,
+            "total": 200
+        },
+        "rm2-tele-far": {
+            "wer": 2.63,
+            "total": 200
+        },
+        "rm3-babb-clo": {
+            "wer": 7.33,
+            "total": 200
+        },
+        "rm3-babb-far": {
+            "wer": 57.48,
+            "total": 200
+        },
+        "rm3-musi-clo": {
+            "wer": 5.02,
+            "total": 200
+        },
+        "rm3-musi-far": {
+            "wer": 37.99,
+            "total": 200
+        },
+        "rm3-none-clo": {
+            "wer": 2.67,
+            "total": 200
+        },
+        "rm3-none-far": {
+            "wer": 12.11,
+            "total": 200
+        },
+        "rm3-tele-clo": {
+            "wer": 5.71,
+            "total": 200
+        },
+        "rm3-tele-far": {
+            "wer": 51.01,
+            "total": 200
+        },
+        "rm4-babb-clo": {
+            "wer": 2.08,
+            "total": 200
+        },
+        "rm4-babb-far": {
+            "wer": 82.05,
+            "total": 200
+        },
+        "rm4-musi-clo": {
+            "wer": 1.41,
+            "total": 200
+        },
+        "rm4-musi-far": {
+            "wer": 13.01,
+            "total": 200
+        },
+        "rm4-none-clo": {
+            "wer": 1.18,
+            "total": 200
+        },
+        "rm4-none-far": {
+            "wer": 2.63,
+            "total": 200
+        },
+        "rm4-tele-clo": {
+            "wer": 1.25,
+            "total": 200
+        },
+        "rm4-tele-far": {
+            "wer": 15.63,
+            "total": 200
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

mini_whole/mini-encoder+align-whole1226_signal_new1-low-lora-1gpu-bs16_1_gckF2000/Step-Audio-2-mini-all-lora3/voices_dev_test/logs/rank7.log

@@ -0,0 +1,2 @@
+2026-01-04 12:23:43 | INFO | Running Step-Audio-2-mini-all-lora3 on dataset: voices_dev_test
+2026-01-04 12:23:43 | INFO | Msg example: {'index': 7, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp1447/Lab41-SRI-VOiCES-rm4-babb-sp1447-ch130551-sg0027-mc05-stu-far-dg140.wav'], 'text': 'Please transcribe the spoken content into written text.', 'meta': {'task': 'ASR', 'interactive': 'Audio-analysis', 'audio_type': 'Speech', 'dataset_series': 'voices_dev', 'dataset_name': 'voices_dev_test', 'lang': 'en', 'subset': 'rm4-babb-far'}}