Upload folder using huggingface_hub

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/Qwen2.5-Omni-7B-encoder-aligner_tedlium_test1_default_performance.json

@@ -0,0 +1,13 @@
+{
+    "task": "ASR",
+    "dataset": "tedlium_test1",
+    "model": "Qwen2.5-Omni-7B-encoder-aligner",
+    "date": "2025-12-17 04:41:46.806676",
+    "performance": {
+        "TEDLIUM-test": {
+            "wer": 4.9,
+            "total": 1155
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank0.log

@@ -0,0 +1,4 @@
+2025-12-17 04:37:16 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:37:16 | 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'}}
+2025-12-17 04:38:56 | INFO | model Qwen2.5-Omni-7B-encoder-aligner, data tedlium_test1, all 8 result merged to 7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/Qwen2.5-Omni-7B-encoder-aligner_tedlium_test1.jsonl.
+2025-12-17 04:38:56 | INFO | skip eval for tedlium_test1

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-17 04:36:20 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:36:20 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-17 04:37:01 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:37:01 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-17 04:36:58 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:36:58 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-17 04:37:03 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:37:03 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-17 04:36:47 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:36:47 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-17 04:36:55 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:36:55 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-17 04:37:10 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:37:10 | 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'}}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/Qwen2.5-Omni-7B-encoder-aligner_voices_dev_test_default_performance.json

@@ -0,0 +1,137 @@
+{
+    "task": "ASR",
+    "dataset": "voices_dev_test",
+    "model": "Qwen2.5-Omni-7B-encoder-aligner",
+    "date": "2025-12-17 04:41:45.771407",
+    "performance": {
+        "rm1-babb-clo": {
+            "wer": 3.38,
+            "total": 200
+        },
+        "rm1-babb-far": {
+            "wer": 6.47,
+            "total": 200
+        },
+        "rm1-musi-clo": {
+            "wer": 3.67,
+            "total": 200
+        },
+        "rm1-musi-far": {
+            "wer": 4.55,
+            "total": 200
+        },
+        "rm1-none-clo": {
+            "wer": 2.63,
+            "total": 200
+        },
+        "rm1-none-far": {
+            "wer": 3.23,
+            "total": 200
+        },
+        "rm1-tele-clo": {
+            "wer": 3.31,
+            "total": 200
+        },
+        "rm1-tele-far": {
+            "wer": 5.75,
+            "total": 200
+        },
+        "rm2-babb-clo": {
+            "wer": 3.39,
+            "total": 200
+        },
+        "rm2-babb-far": {
+            "wer": 6.64,
+            "total": 200
+        },
+        "rm2-musi-clo": {
+            "wer": 3.32,
+            "total": 200
+        },
+        "rm2-musi-far": {
+            "wer": 4.37,
+            "total": 200
+        },
+        "rm2-none-clo": {
+            "wer": 2.73,
+            "total": 200
+        },
+        "rm2-none-far": {
+            "wer": 3.38,
+            "total": 200
+        },
+        "rm2-tele-clo": {
+            "wer": 3.41,
+            "total": 200
+        },
+        "rm2-tele-far": {
+            "wer": 6.03,
+            "total": 200
+        },
+        "rm3-babb-clo": {
+            "wer": 14.87,
+            "total": 200
+        },
+        "rm3-babb-far": {
+            "wer": 82.13,
+            "total": 200
+        },
+        "rm3-musi-clo": {
+            "wer": 10.86,
+            "total": 200
+        },
+        "rm3-musi-far": {
+            "wer": 58.57,
+            "total": 200
+        },
+        "rm3-none-clo": {
+            "wer": 5.22,
+            "total": 200
+        },
+        "rm3-none-far": {
+            "wer": 22.84,
+            "total": 200
+        },
+        "rm3-tele-clo": {
+            "wer": 10.8,
+            "total": 200
+        },
+        "rm3-tele-far": {
+            "wer": 65.42,
+            "total": 200
+        },
+        "rm4-babb-clo": {
+            "wer": 4.17,
+            "total": 200
+        },
+        "rm4-babb-far": {
+            "wer": 90.53,
+            "total": 200
+        },
+        "rm4-musi-clo": {
+            "wer": 3.07,
+            "total": 200
+        },
+        "rm4-musi-far": {
+            "wer": 26.59,
+            "total": 200
+        },
+        "rm4-none-clo": {
+            "wer": 2.67,
+            "total": 200
+        },
+        "rm4-none-far": {
+            "wer": 5.04,
+            "total": 200
+        },
+        "rm4-tele-clo": {
+            "wer": 3.17,
+            "total": 200
+        },
+        "rm4-tele-far": {
+            "wer": 26.57,
+            "total": 200
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank0.log

@@ -0,0 +1,8 @@
+2025-12-17 04:22:45 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:45 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.
+2025-12-17 04:37:06 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-17 04:37:16 | INFO | waiting for other ranks to finish, time elapsed: 20s
+2025-12-17 04:37:16 | INFO | model Qwen2.5-Omni-7B-encoder-aligner, data voices_dev_test, all 8 result merged to 7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/Qwen2.5-Omni-7B-encoder-aligner_voices_dev_test.jsonl.
+2025-12-17 04:37:16 | INFO | skip eval for voices_dev_test

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank1.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:44 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:44 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank2.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:44 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:44 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank3.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:44 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:44 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank4.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:44 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:44 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank5.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:43 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:43 | 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'}}
+2025-12-17 04:22:44 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank6.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:42 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:42 | 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'}}
+2025-12-17 04:22:43 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

7B_ea_ff_linear_ck600/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank7.log

@@ -0,0 +1,4 @@
+2025-12-17 04:22:44 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:22:44 | 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'}}
+2025-12-17 04:22:45 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.