Add files using upload-large-folder tool

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/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 05:48:31.723212",
+    "performance": {
+        "TEDLIUM-test": {
+            "wer": 5.04,
+            "total": 1155
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank0.log

@@ -0,0 +1,4 @@
+2025-12-17 05:21:53 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:53 | 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 05:23:32 | INFO | model Qwen2.5-Omni-7B-encoder-aligner, data tedlium_test1, all 8 result merged to 7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/Qwen2.5-Omni-7B-encoder-aligner_tedlium_test1.jsonl.
+2025-12-17 05:23:32 | INFO | skip eval for tedlium_test1

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:12 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:12 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:45 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:45 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:43 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:43 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:22 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:22 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:34 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:34 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:26 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:26 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-17 05:21:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 05:21:48 | 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'}}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/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 05:48:30.681045",
+    "performance": {
+        "rm1-babb-clo": {
+            "wer": 4.06,
+            "total": 200
+        },
+        "rm1-babb-far": {
+            "wer": 6.63,
+            "total": 200
+        },
+        "rm1-musi-clo": {
+            "wer": 3.78,
+            "total": 200
+        },
+        "rm1-musi-far": {
+            "wer": 4.7,
+            "total": 200
+        },
+        "rm1-none-clo": {
+            "wer": 2.71,
+            "total": 200
+        },
+        "rm1-none-far": {
+            "wer": 3.46,
+            "total": 200
+        },
+        "rm1-tele-clo": {
+            "wer": 3.47,
+            "total": 200
+        },
+        "rm1-tele-far": {
+            "wer": 5.57,
+            "total": 200
+        },
+        "rm2-babb-clo": {
+            "wer": 3.92,
+            "total": 200
+        },
+        "rm2-babb-far": {
+            "wer": 7.4,
+            "total": 200
+        },
+        "rm2-musi-clo": {
+            "wer": 3.41,
+            "total": 200
+        },
+        "rm2-musi-far": {
+            "wer": 4.32,
+            "total": 200
+        },
+        "rm2-none-clo": {
+            "wer": 2.96,
+            "total": 200
+        },
+        "rm2-none-far": {
+            "wer": 3.41,
+            "total": 200
+        },
+        "rm2-tele-clo": {
+            "wer": 3.7,
+            "total": 200
+        },
+        "rm2-tele-far": {
+            "wer": 6.12,
+            "total": 200
+        },
+        "rm3-babb-clo": {
+            "wer": 15.01,
+            "total": 200
+        },
+        "rm3-babb-far": {
+            "wer": 75.95,
+            "total": 200
+        },
+        "rm3-musi-clo": {
+            "wer": 11.17,
+            "total": 200
+        },
+        "rm3-musi-far": {
+            "wer": 58.43,
+            "total": 200
+        },
+        "rm3-none-clo": {
+            "wer": 5.23,
+            "total": 200
+        },
+        "rm3-none-far": {
+            "wer": 23.94,
+            "total": 200
+        },
+        "rm3-tele-clo": {
+            "wer": 11.05,
+            "total": 200
+        },
+        "rm3-tele-far": {
+            "wer": 77.17,
+            "total": 200
+        },
+        "rm4-babb-clo": {
+            "wer": 4.35,
+            "total": 200
+        },
+        "rm4-babb-far": {
+            "wer": 83.5,
+            "total": 200
+        },
+        "rm4-musi-clo": {
+            "wer": 3.15,
+            "total": 200
+        },
+        "rm4-musi-far": {
+            "wer": 26.67,
+            "total": 200
+        },
+        "rm4-none-clo": {
+            "wer": 2.91,
+            "total": 200
+        },
+        "rm4-none-far": {
+            "wer": 5.13,
+            "total": 200
+        },
+        "rm4-tele-clo": {
+            "wer": 3.25,
+            "total": 200
+        },
+        "rm4-tele-far": {
+            "wer": 28.05,
+            "total": 200
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank3.log

@@ -0,0 +1,4 @@
+2025-12-17 05:07:34 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 05:07:34 | 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 05:07:35 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1200/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank7.log

@@ -0,0 +1,4 @@
+2025-12-17 05:07:35 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 05:07:35 | 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 05:07:35 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck1500/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-17 06:04:15 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 06:04:15 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck1500/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-17 06:04:41 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 06:04:41 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck300/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 04:19:43.964241",
+    "performance": {
+        "TEDLIUM-test": {
+            "wer": 4.59,
+            "total": 1155
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank0.log

@@ -0,0 +1,7 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:48 | 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 04:15:49 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.
+2025-12-17 04:17:40 | INFO | waiting for other ranks to finish, time elapsed: 10s
+2025-12-17 04:17:40 | INFO | model Qwen2.5-Omni-7B-encoder-aligner, data tedlium_test1, all 8 result merged to 7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/Qwen2.5-Omni-7B-encoder-aligner_tedlium_test1.jsonl.
+2025-12-17 04:17:40 | INFO | skip eval for tedlium_test1

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank1.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:48 | 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'}}
+2025-12-17 04:15:48 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank2.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:49 | 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'}}
+2025-12-17 04:15:50 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank3.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:50 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:50 | 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'}}
+2025-12-17 04:15:51 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank4.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:49 | 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'}}
+2025-12-17 04:15:50 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank5.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:49 | 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'}}
+2025-12-17 04:15:50 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank6.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:49 | 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'}}
+2025-12-17 04:15:50 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank7.log

@@ -0,0 +1,4 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:15:48 | 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'}}
+2025-12-17 04:15:49 | INFO | Prompt: You are a speech recognition model.
+Transcribe the English audio into text without any punctuation marks.

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/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 04:19:42.949893",
+    "performance": {
+        "rm1-babb-clo": {
+            "wer": 3.24,
+            "total": 200
+        },
+        "rm1-babb-far": {
+            "wer": 5.73,
+            "total": 200
+        },
+        "rm1-musi-clo": {
+            "wer": 3.22,
+            "total": 200
+        },
+        "rm1-musi-far": {
+            "wer": 4.46,
+            "total": 200
+        },
+        "rm1-none-clo": {
+            "wer": 2.43,
+            "total": 200
+        },
+        "rm1-none-far": {
+            "wer": 2.94,
+            "total": 200
+        },
+        "rm1-tele-clo": {
+            "wer": 3.12,
+            "total": 200
+        },
+        "rm1-tele-far": {
+            "wer": 5.23,
+            "total": 200
+        },
+        "rm2-babb-clo": {
+            "wer": 3.29,
+            "total": 200
+        },
+        "rm2-babb-far": {
+            "wer": 6.33,
+            "total": 200
+        },
+        "rm2-musi-clo": {
+            "wer": 2.96,
+            "total": 200
+        },
+        "rm2-musi-far": {
+            "wer": 4.59,
+            "total": 200
+        },
+        "rm2-none-clo": {
+            "wer": 2.58,
+            "total": 200
+        },
+        "rm2-none-far": {
+            "wer": 3.3,
+            "total": 200
+        },
+        "rm2-tele-clo": {
+            "wer": 3.19,
+            "total": 200
+        },
+        "rm2-tele-far": {
+            "wer": 5.52,
+            "total": 200
+        },
+        "rm3-babb-clo": {
+            "wer": 14.11,
+            "total": 200
+        },
+        "rm3-babb-far": {
+            "wer": 96.84,
+            "total": 200
+        },
+        "rm3-musi-clo": {
+            "wer": 10.11,
+            "total": 200
+        },
+        "rm3-musi-far": {
+            "wer": 69.01,
+            "total": 200
+        },
+        "rm3-none-clo": {
+            "wer": 4.87,
+            "total": 200
+        },
+        "rm3-none-far": {
+            "wer": 22.82,
+            "total": 200
+        },
+        "rm3-tele-clo": {
+            "wer": 10.75,
+            "total": 200
+        },
+        "rm3-tele-far": {
+            "wer": 78.05,
+            "total": 200
+        },
+        "rm4-babb-clo": {
+            "wer": 3.83,
+            "total": 200
+        },
+        "rm4-babb-far": {
+            "wer": 105.91,
+            "total": 200
+        },
+        "rm4-musi-clo": {
+            "wer": 2.85,
+            "total": 200
+        },
+        "rm4-musi-far": {
+            "wer": 23.0,
+            "total": 200
+        },
+        "rm4-none-clo": {
+            "wer": 2.53,
+            "total": 200
+        },
+        "rm4-none-far": {
+            "wer": 4.5,
+            "total": 200
+        },
+        "rm4-tele-clo": {
+            "wer": 2.82,
+            "total": 200
+        },
+        "rm4-tele-far": {
+            "wer": 29.6,
+            "total": 200
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank0.log

@@ -0,0 +1,3 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test
+2025-12-17 04:15:48 | INFO | file 7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/Qwen2.5-Omni-7B-encoder-aligner_voices_dev_test.jsonl exists, reevaluating...
+2025-12-17 04:15:48 | INFO | skip eval for voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank1.log

@@ -0,0 +1 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank2.log

@@ -0,0 +1 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank3.log

@@ -0,0 +1 @@
+2025-12-17 04:15:50 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank4.log

@@ -0,0 +1 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank5.log

@@ -0,0 +1 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank6.log

@@ -0,0 +1 @@
+2025-12-17 04:15:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck300/Qwen2.5-Omni-7B-encoder-aligner/voices_dev_test/logs/rank7.log

@@ -0,0 +1 @@
+2025-12-17 04:15:48 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: voices_dev_test

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck900/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 05:06:22.190944",
+    "performance": {
+        "TEDLIUM-test": {
+            "wer": 4.97,
+            "total": 1155
+        }
+    },
+    "eval_method": "qwen2-audio-impl"
+}

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank0.log

@@ -0,0 +1,4 @@
+2025-12-17 04:57:56 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:56 | 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 04:59:35 | INFO | model Qwen2.5-Omni-7B-encoder-aligner, data tedlium_test1, all 8 result merged to 7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/Qwen2.5-Omni-7B-encoder-aligner_tedlium_test1.jsonl.
+2025-12-17 04:59:35 | INFO | skip eval for tedlium_test1

different_distribution/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank1.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:12 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:12 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank2.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:23 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:23 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank3.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:49 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:49 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank4.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:19 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:19 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank5.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:16 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:16 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank6.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:11 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:11 | 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/7B_ff_ea_linear/7B_ea_ff_linear_ck900/Qwen2.5-Omni-7B-encoder-aligner/tedlium_test1/logs/rank7.log

@@ -0,0 +1,2 @@
+2025-12-17 04:57:34 | INFO | Running Qwen2.5-Omni-7B-encoder-aligner on dataset: tedlium_test1
+2025-12-17 04:57:34 | 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'}}