Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1_default_performance.json +13 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1_wer_details.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank0.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank1.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank2.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank3.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank4.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank5.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank6.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank7.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank1.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank2.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank3.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank4.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank5.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank6.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank7.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1_default_performance.json +13 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1_wer_details.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank0.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank1.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank2.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank3.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank4.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank5.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank6.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank7.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test_default_performance.json +137 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test_wer_details.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank0.log +11 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank1.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank2.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank3.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank4.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank5.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank6.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank7.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1_default_performance.json +13 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1_wer_details.jsonl +0 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank0.log +4 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank1.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank2.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank3.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank4.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank5.log +2 -0
- different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank6.log +2 -0
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1_default_performance.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"task": "ASR",
|
| 3 |
+
"dataset": "tedlium_test1",
|
| 4 |
+
"model": "Qwen2.5-Omni-7B-lora1-ff",
|
| 5 |
+
"date": "2025-12-21 12:16:47.542656",
|
| 6 |
+
"performance": {
|
| 7 |
+
"TEDLIUM-test": {
|
| 8 |
+
"wer": 4.74,
|
| 9 |
+
"total": 1155
|
| 10 |
+
}
|
| 11 |
+
},
|
| 12 |
+
"eval_method": "qwen2-audio-impl"
|
| 13 |
+
}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1_wer_details.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank0.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:49 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:49 | 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'}}
|
| 3 |
+
2025-12-21 06:42:16 | INFO | model Qwen2.5-Omni-7B-lora1-ff, data tedlium_test1, all 8 result merged to ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/Qwen2.5-Omni-7B-lora1-ff_tedlium_test1.jsonl.
|
| 4 |
+
2025-12-21 06:42:16 | INFO | skip eval for tedlium_test1
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank1.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:33 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:33 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank2.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:26 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:26 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank3.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:33 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:33 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank4.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:33 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:33 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank5.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:30 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:30 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank6.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:43 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:43 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/tedlium_test1/logs/rank7.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:40:29 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:40:29 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank1.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:11 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:11 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032639-sg0029-mc05-stu-far-dg160.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'}}
|
| 3 |
+
2025-12-21 06:28:12 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank2.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:15 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:15 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032658-sg0012-mc05-stu-far-dg070.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'}}
|
| 3 |
+
2025-12-21 06:28:16 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank3.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:13 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:13 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032658-sg0012-mc01-stu-clo-dg070.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'}}
|
| 3 |
+
2025-12-21 06:28:14 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank4.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:16 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:16 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp1447/Lab41-SRI-VOiCES-rm4-babb-sp1447-ch130550-sg0026-mc01-stu-clo-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-clo'}}
|
| 3 |
+
2025-12-21 06:28:16 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank5.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:13 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:13 | 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'}}
|
| 3 |
+
2025-12-21 06:28:14 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank6.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:13 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:13 | 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'}}
|
| 3 |
+
2025-12-21 06:28:14 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora1-ff/voices_dev_test/logs/rank7.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:28:16 | INFO | Running Qwen2.5-Omni-7B-lora1-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:28:16 | 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'}}
|
| 3 |
+
2025-12-21 06:28:16 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1_default_performance.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"task": "ASR",
|
| 3 |
+
"dataset": "tedlium_test1",
|
| 4 |
+
"model": "Qwen2.5-Omni-7B-lora2-ff",
|
| 5 |
+
"date": "2025-12-21 12:17:27.230158",
|
| 6 |
+
"performance": {
|
| 7 |
+
"TEDLIUM-test": {
|
| 8 |
+
"wer": 4.42,
|
| 9 |
+
"total": 1155
|
| 10 |
+
}
|
| 11 |
+
},
|
| 12 |
+
"eval_method": "qwen2-audio-impl"
|
| 13 |
+
}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1_wer_details.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank0.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:56:24 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:56:24 | 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'}}
|
| 3 |
+
2025-12-21 06:57:48 | INFO | model Qwen2.5-Omni-7B-lora2-ff, data tedlium_test1, all 8 result merged to ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/Qwen2.5-Omni-7B-lora2-ff_tedlium_test1.jsonl.
|
| 4 |
+
2025-12-21 06:57:48 | INFO | skip eval for tedlium_test1
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank1.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:55:43 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:55:43 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank2.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:55:35 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:55:35 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank3.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:56:03 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:56:03 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank4.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:55:45 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:55:45 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank5.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:55:53 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:55:53 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank6.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:56:14 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:56:14 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/tedlium_test1/logs/rank7.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:56:02 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 06:56:02 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test_default_performance.json
ADDED
|
@@ -0,0 +1,137 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"task": "ASR",
|
| 3 |
+
"dataset": "voices_dev_test",
|
| 4 |
+
"model": "Qwen2.5-Omni-7B-lora2-ff",
|
| 5 |
+
"date": "2025-12-21 12:17:26.477078",
|
| 6 |
+
"performance": {
|
| 7 |
+
"rm1-babb-clo": {
|
| 8 |
+
"wer": 3.26,
|
| 9 |
+
"total": 200
|
| 10 |
+
},
|
| 11 |
+
"rm1-babb-far": {
|
| 12 |
+
"wer": 5.86,
|
| 13 |
+
"total": 200
|
| 14 |
+
},
|
| 15 |
+
"rm1-musi-clo": {
|
| 16 |
+
"wer": 3.22,
|
| 17 |
+
"total": 200
|
| 18 |
+
},
|
| 19 |
+
"rm1-musi-far": {
|
| 20 |
+
"wer": 4.47,
|
| 21 |
+
"total": 200
|
| 22 |
+
},
|
| 23 |
+
"rm1-none-clo": {
|
| 24 |
+
"wer": 2.56,
|
| 25 |
+
"total": 200
|
| 26 |
+
},
|
| 27 |
+
"rm1-none-far": {
|
| 28 |
+
"wer": 3.15,
|
| 29 |
+
"total": 200
|
| 30 |
+
},
|
| 31 |
+
"rm1-tele-clo": {
|
| 32 |
+
"wer": 3.4,
|
| 33 |
+
"total": 200
|
| 34 |
+
},
|
| 35 |
+
"rm1-tele-far": {
|
| 36 |
+
"wer": 5.28,
|
| 37 |
+
"total": 200
|
| 38 |
+
},
|
| 39 |
+
"rm2-babb-clo": {
|
| 40 |
+
"wer": 3.31,
|
| 41 |
+
"total": 200
|
| 42 |
+
},
|
| 43 |
+
"rm2-babb-far": {
|
| 44 |
+
"wer": 6.24,
|
| 45 |
+
"total": 200
|
| 46 |
+
},
|
| 47 |
+
"rm2-musi-clo": {
|
| 48 |
+
"wer": 3.15,
|
| 49 |
+
"total": 200
|
| 50 |
+
},
|
| 51 |
+
"rm2-musi-far": {
|
| 52 |
+
"wer": 4.62,
|
| 53 |
+
"total": 200
|
| 54 |
+
},
|
| 55 |
+
"rm2-none-clo": {
|
| 56 |
+
"wer": 2.73,
|
| 57 |
+
"total": 200
|
| 58 |
+
},
|
| 59 |
+
"rm2-none-far": {
|
| 60 |
+
"wer": 3.41,
|
| 61 |
+
"total": 200
|
| 62 |
+
},
|
| 63 |
+
"rm2-tele-clo": {
|
| 64 |
+
"wer": 3.39,
|
| 65 |
+
"total": 200
|
| 66 |
+
},
|
| 67 |
+
"rm2-tele-far": {
|
| 68 |
+
"wer": 5.57,
|
| 69 |
+
"total": 200
|
| 70 |
+
},
|
| 71 |
+
"rm3-babb-clo": {
|
| 72 |
+
"wer": 14.25,
|
| 73 |
+
"total": 200
|
| 74 |
+
},
|
| 75 |
+
"rm3-babb-far": {
|
| 76 |
+
"wer": 98.18,
|
| 77 |
+
"total": 200
|
| 78 |
+
},
|
| 79 |
+
"rm3-musi-clo": {
|
| 80 |
+
"wer": 9.88,
|
| 81 |
+
"total": 200
|
| 82 |
+
},
|
| 83 |
+
"rm3-musi-far": {
|
| 84 |
+
"wer": 69.47,
|
| 85 |
+
"total": 200
|
| 86 |
+
},
|
| 87 |
+
"rm3-none-clo": {
|
| 88 |
+
"wer": 4.75,
|
| 89 |
+
"total": 200
|
| 90 |
+
},
|
| 91 |
+
"rm3-none-far": {
|
| 92 |
+
"wer": 23.37,
|
| 93 |
+
"total": 200
|
| 94 |
+
},
|
| 95 |
+
"rm3-tele-clo": {
|
| 96 |
+
"wer": 10.68,
|
| 97 |
+
"total": 200
|
| 98 |
+
},
|
| 99 |
+
"rm3-tele-far": {
|
| 100 |
+
"wer": 81.51,
|
| 101 |
+
"total": 200
|
| 102 |
+
},
|
| 103 |
+
"rm4-babb-clo": {
|
| 104 |
+
"wer": 3.99,
|
| 105 |
+
"total": 200
|
| 106 |
+
},
|
| 107 |
+
"rm4-babb-far": {
|
| 108 |
+
"wer": 112.28,
|
| 109 |
+
"total": 200
|
| 110 |
+
},
|
| 111 |
+
"rm4-musi-clo": {
|
| 112 |
+
"wer": 2.95,
|
| 113 |
+
"total": 200
|
| 114 |
+
},
|
| 115 |
+
"rm4-musi-far": {
|
| 116 |
+
"wer": 25.8,
|
| 117 |
+
"total": 200
|
| 118 |
+
},
|
| 119 |
+
"rm4-none-clo": {
|
| 120 |
+
"wer": 2.54,
|
| 121 |
+
"total": 200
|
| 122 |
+
},
|
| 123 |
+
"rm4-none-far": {
|
| 124 |
+
"wer": 4.62,
|
| 125 |
+
"total": 200
|
| 126 |
+
},
|
| 127 |
+
"rm4-tele-clo": {
|
| 128 |
+
"wer": 2.91,
|
| 129 |
+
"total": 200
|
| 130 |
+
},
|
| 131 |
+
"rm4-tele-far": {
|
| 132 |
+
"wer": 24.89,
|
| 133 |
+
"total": 200
|
| 134 |
+
}
|
| 135 |
+
},
|
| 136 |
+
"eval_method": "qwen2-audio-impl"
|
| 137 |
+
}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test_wer_details.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank0.log
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:38 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:38 | INFO | Msg example: {'index': 0, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032639-sg0029-mc01-stu-clo-dg160.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'}}
|
| 3 |
+
2025-12-21 06:43:38 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
| 5 |
+
2025-12-21 06:55:44 | INFO | waiting for other ranks to finish, time elapsed: 10s
|
| 6 |
+
2025-12-21 06:55:54 | INFO | waiting for other ranks to finish, time elapsed: 20s
|
| 7 |
+
2025-12-21 06:56:04 | INFO | waiting for other ranks to finish, time elapsed: 30s
|
| 8 |
+
2025-12-21 06:56:14 | INFO | waiting for other ranks to finish, time elapsed: 40s
|
| 9 |
+
2025-12-21 06:56:24 | INFO | waiting for other ranks to finish, time elapsed: 50s
|
| 10 |
+
2025-12-21 06:56:24 | INFO | model Qwen2.5-Omni-7B-lora2-ff, data voices_dev_test, all 8 result merged to ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/Qwen2.5-Omni-7B-lora2-ff_voices_dev_test.jsonl.
|
| 11 |
+
2025-12-21 06:56:24 | INFO | skip eval for voices_dev_test
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank1.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:37 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:37 | INFO | Msg example: {'index': 1, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032639-sg0029-mc05-stu-far-dg160.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'}}
|
| 3 |
+
2025-12-21 06:43:37 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank2.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:36 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:36 | INFO | Msg example: {'index': 2, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032658-sg0012-mc05-stu-far-dg070.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'}}
|
| 3 |
+
2025-12-21 06:43:37 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank3.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:38 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:38 | INFO | Msg example: {'index': 3, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp4899/Lab41-SRI-VOiCES-rm4-babb-sp4899-ch032658-sg0012-mc01-stu-clo-dg070.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'}}
|
| 3 |
+
2025-12-21 06:43:38 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank4.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:38 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:38 | INFO | Msg example: {'index': 4, 'audio': ['/workspace/intern/pangkaiyu/dg/VOiCES_devkit/distant-16k/speech/test/rm4/babb/sp1447/Lab41-SRI-VOiCES-rm4-babb-sp1447-ch130550-sg0026-mc01-stu-clo-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-clo'}}
|
| 3 |
+
2025-12-21 06:43:38 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank5.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:38 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:38 | 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'}}
|
| 3 |
+
2025-12-21 06:43:39 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank6.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:41 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:41 | 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'}}
|
| 3 |
+
2025-12-21 06:43:41 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora2-ff/voices_dev_test/logs/rank7.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 06:43:36 | INFO | Running Qwen2.5-Omni-7B-lora2-ff on dataset: voices_dev_test
|
| 2 |
+
2025-12-21 06:43:36 | 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'}}
|
| 3 |
+
2025-12-21 06:43:37 | INFO | Prompt: You are a speech recognition model.
|
| 4 |
+
Transcribe the English audio into text without any punctuation marks.
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1_default_performance.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"task": "ASR",
|
| 3 |
+
"dataset": "tedlium_test1",
|
| 4 |
+
"model": "Qwen2.5-Omni-7B-lora3-ff",
|
| 5 |
+
"date": "2025-12-21 12:18:06.065987",
|
| 6 |
+
"performance": {
|
| 7 |
+
"TEDLIUM-test": {
|
| 8 |
+
"wer": 4.42,
|
| 9 |
+
"total": 1155
|
| 10 |
+
}
|
| 11 |
+
},
|
| 12 |
+
"eval_method": "qwen2-audio-impl"
|
| 13 |
+
}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1_wer_details.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank0.log
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:11:24 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:11:24 | 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'}}
|
| 3 |
+
2025-12-21 07:12:48 | INFO | model Qwen2.5-Omni-7B-lora3-ff, data tedlium_test1, all 8 result merged to ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/Qwen2.5-Omni-7B-lora3-ff_tedlium_test1.jsonl.
|
| 4 |
+
2025-12-21 07:12:48 | INFO | skip eval for tedlium_test1
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank1.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:51 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:51 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank2.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:57 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:57 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank3.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:51 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:51 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank4.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:48 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:48 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank5.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:52 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:52 | 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'}}
|
different_distribution/ff_5e-6_gaussion/Qwen2.5-Omni-7B-lora3-ff/tedlium_test1/logs/rank6.log
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-12-21 07:10:55 | INFO | Running Qwen2.5-Omni-7B-lora3-ff on dataset: tedlium_test1
|
| 2 |
+
2025-12-21 07:10:55 | 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'}}
|