update model card README.md
Browse files- README.md +31 -2
- fine-tune-whisper-streaming.ipynb +3 -29
README.md
CHANGED
@@ -1,15 +1,30 @@
|
|
1 |
---
|
2 |
language:
|
3 |
-
- zh
|
4 |
license: apache-2.0
|
5 |
tags:
|
6 |
- whisper-event
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- mozilla-foundation/common_voice_11_0
|
|
|
|
|
10 |
model-index:
|
11 |
- name: Whisper Small Chinese - Kimbo Chen
|
12 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
---
|
14 |
|
15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -18,6 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
# Whisper Small Chinese - Kimbo Chen
|
19 |
|
20 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
|
|
|
|
|
|
|
21 |
|
22 |
## Model description
|
23 |
|
@@ -46,6 +64,17 @@ The following hyperparameters were used during training:
|
|
46 |
- training_steps: 5000
|
47 |
- mixed_precision_training: Native AMP
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
### Framework versions
|
50 |
|
51 |
- Transformers 4.26.0.dev0
|
|
|
1 |
---
|
2 |
language:
|
3 |
+
- zh-TW
|
4 |
license: apache-2.0
|
5 |
tags:
|
6 |
- whisper-event
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- mozilla-foundation/common_voice_11_0
|
10 |
+
metrics:
|
11 |
+
- wer
|
12 |
model-index:
|
13 |
- name: Whisper Small Chinese - Kimbo Chen
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Automatic Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: Common Voice 11.0
|
20 |
+
type: mozilla-foundation/common_voice_11_0
|
21 |
+
config: zh-TW
|
22 |
+
split: test
|
23 |
+
args: zh-TW
|
24 |
+
metrics:
|
25 |
+
- name: Wer
|
26 |
+
type: wer
|
27 |
+
value: 45.8546571136131
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
33 |
# Whisper Small Chinese - Kimbo Chen
|
34 |
|
35 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
|
36 |
+
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.2846
|
38 |
+
- Wer: 45.8547
|
39 |
|
40 |
## Model description
|
41 |
|
|
|
64 |
- training_steps: 5000
|
65 |
- mixed_precision_training: Native AMP
|
66 |
|
67 |
+
### Training results
|
68 |
+
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
70 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
71 |
+
| 0.0016 | 9.01 | 1000 | 0.2376 | 44.4422 |
|
72 |
+
| 0.0004 | 19.01 | 2000 | 0.2559 | 44.6673 |
|
73 |
+
| 0.0002 | 29.0 | 3000 | 0.2710 | 45.2815 |
|
74 |
+
| 0.0002 | 38.02 | 4000 | 0.2804 | 45.7523 |
|
75 |
+
| 0.0001 | 48.01 | 5000 | 0.2846 | 45.8547 |
|
76 |
+
|
77 |
+
|
78 |
### Framework versions
|
79 |
|
80 |
- Transformers 4.26.0.dev0
|
fine-tune-whisper-streaming.ipynb
CHANGED
@@ -1141,7 +1141,7 @@
|
|
1141 |
},
|
1142 |
{
|
1143 |
"cell_type": "code",
|
1144 |
-
"execution_count":
|
1145 |
"id": "95737cda-c5dd-4887-a4d0-dfcb0d61d977",
|
1146 |
"metadata": {},
|
1147 |
"outputs": [
|
@@ -1156,40 +1156,14 @@
|
|
1156 |
"tokenizer config file saved in ./tokenizer_config.json\n",
|
1157 |
"Special tokens file saved in ./special_tokens_map.json\n",
|
1158 |
"added tokens file saved in ./added_tokens.json\n",
|
1159 |
-
"Several commits (
|
1160 |
"The progress bars may be unreliable.\n",
|
1161 |
"remote: Scanning LFS files for validity, may be slow... \n",
|
1162 |
"remote: LFS file scan complete. \n",
|
1163 |
-
"remote: ----------------------------------------------------------\u001b[0;31m \n",
|
1164 |
-
"remote: Sorry, your push was rejected during YAML metadata verification: \n",
|
1165 |
-
"remote: - Error: \"language[0]\" with value \"zh-tw\" is not valid. It must be an ISO 639-1, 639-2 or 639-3 code (two/three letters), or a special value like \"code\", \"multilingual\". If you want to use BCP-47 identifiers, you can specify them in language_bcp47.\u001b[0;32m \n",
|
1166 |
-
"remote: ---------------------------------------------------------- \n",
|
1167 |
-
"remote: Please find the documentation at: \n",
|
1168 |
-
"remote: https://huggingface.co/docs/hub/model-cards#model-card-metadata\u001b[0;0m \n",
|
1169 |
-
"remote: ---------------------------------------------------------- \n",
|
1170 |
"To https://huggingface.co/kimbochen/whisper-small-zh-tw\n",
|
1171 |
-
"
|
1172 |
-
"error: failed to push some refs to 'https://huggingface.co/kimbochen/whisper-small-zh-tw'\n",
|
1173 |
"\n"
|
1174 |
]
|
1175 |
-
},
|
1176 |
-
{
|
1177 |
-
"ename": "OSError",
|
1178 |
-
"evalue": "remote: Scanning LFS files for validity, may be slow... \nremote: LFS file scan complete. \nremote: ----------------------------------------------------------\u001b[0;31m \nremote: Sorry, your push was rejected during YAML metadata verification: \nremote: - Error: \"language[0]\" with value \"zh-tw\" is not valid. It must be an ISO 639-1, 639-2 or 639-3 code (two/three letters), or a special value like \"code\", \"multilingual\". If you want to use BCP-47 identifiers, you can specify them in language_bcp47.\u001b[0;32m \nremote: ---------------------------------------------------------- \nremote: Please find the documentation at: \nremote: https://huggingface.co/docs/hub/model-cards#model-card-metadata\u001b[0;0m \nremote: ---------------------------------------------------------- \nTo https://huggingface.co/kimbochen/whisper-small-zh-tw\n ! [remote rejected] main -> main (pre-receive hook declined)\nerror: failed to push some refs to 'https://huggingface.co/kimbochen/whisper-small-zh-tw'\n",
|
1179 |
-
"output_type": "error",
|
1180 |
-
"traceback": [
|
1181 |
-
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
|
1182 |
-
"\u001b[0;31mCalledProcessError\u001b[0m Traceback (most recent call last)",
|
1183 |
-
"File \u001b[0;32m~/.venv/lib/python3.8/site-packages/huggingface_hub/repository.py:1207\u001b[0m, in \u001b[0;36mRepository.git_push\u001b[0;34m(self, upstream, blocking, auto_lfs_prune)\u001b[0m\n\u001b[1;32m 1206\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m return_code:\n\u001b[0;32m-> 1207\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m subprocess\u001b[38;5;241m.\u001b[39mCalledProcessError(\n\u001b[1;32m 1208\u001b[0m return_code, process\u001b[38;5;241m.\u001b[39margs, output\u001b[38;5;241m=\u001b[39mstdout, stderr\u001b[38;5;241m=\u001b[39mstderr\n\u001b[1;32m 1209\u001b[0m )\n\u001b[1;32m 1211\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m subprocess\u001b[38;5;241m.\u001b[39mCalledProcessError \u001b[38;5;28;01mas\u001b[39;00m exc:\n",
|
1184 |
-
"\u001b[0;31mCalledProcessError\u001b[0m: Command '['git', 'push', '--set-upstream', 'origin', 'main']' returned non-zero exit status 1.",
|
1185 |
-
"\nDuring handling of the above exception, another exception occurred:\n",
|
1186 |
-
"\u001b[0;31mOSError\u001b[0m Traceback (most recent call last)",
|
1187 |
-
"Cell \u001b[0;32mIn[36], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mtrainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpush_to_hub\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
|
1188 |
-
"File \u001b[0;32m~/.venv/lib/python3.8/site-packages/transformers/trainer.py:3492\u001b[0m, in \u001b[0;36mTrainer.push_to_hub\u001b[0;34m(self, commit_message, blocking, **kwargs)\u001b[0m\n\u001b[1;32m 3489\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpush_in_progress\u001b[38;5;241m.\u001b[39m_process\u001b[38;5;241m.\u001b[39mkill()\n\u001b[1;32m 3490\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpush_in_progress \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m-> 3492\u001b[0m git_head_commit_url \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrepo\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpush_to_hub\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 3493\u001b[0m \u001b[43m \u001b[49m\u001b[43mcommit_message\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_message\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mblocking\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mblocking\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mauto_lfs_prune\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\n\u001b[1;32m 3494\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 3495\u001b[0m \u001b[38;5;66;03m# push separately the model card to be independant from the rest of the model\u001b[39;00m\n\u001b[1;32m 3496\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mshould_save:\n",
|
1189 |
-
"File \u001b[0;32m~/.venv/lib/python3.8/site-packages/huggingface_hub/repository.py:1432\u001b[0m, in \u001b[0;36mRepository.push_to_hub\u001b[0;34m(self, commit_message, blocking, clean_ok, auto_lfs_prune)\u001b[0m\n\u001b[1;32m 1430\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mgit_add(auto_lfs_track\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m 1431\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mgit_commit(commit_message)\n\u001b[0;32m-> 1432\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgit_push\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 1433\u001b[0m \u001b[43m \u001b[49m\u001b[43mupstream\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43mf\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43morigin \u001b[39;49m\u001b[38;5;132;43;01m{\u001b[39;49;00m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcurrent_branch\u001b[49m\u001b[38;5;132;43;01m}\u001b[39;49;00m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m 1434\u001b[0m \u001b[43m \u001b[49m\u001b[43mblocking\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mblocking\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 1435\u001b[0m \u001b[43m \u001b[49m\u001b[43mauto_lfs_prune\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mauto_lfs_prune\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 1436\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
|
1190 |
-
"File \u001b[0;32m~/.venv/lib/python3.8/site-packages/huggingface_hub/repository.py:1212\u001b[0m, in \u001b[0;36mRepository.git_push\u001b[0;34m(self, upstream, blocking, auto_lfs_prune)\u001b[0m\n\u001b[1;32m 1207\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m subprocess\u001b[38;5;241m.\u001b[39mCalledProcessError(\n\u001b[1;32m 1208\u001b[0m return_code, process\u001b[38;5;241m.\u001b[39margs, output\u001b[38;5;241m=\u001b[39mstdout, stderr\u001b[38;5;241m=\u001b[39mstderr\n\u001b[1;32m 1209\u001b[0m )\n\u001b[1;32m 1211\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m subprocess\u001b[38;5;241m.\u001b[39mCalledProcessError \u001b[38;5;28;01mas\u001b[39;00m exc:\n\u001b[0;32m-> 1212\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mEnvironmentError\u001b[39;00m(exc\u001b[38;5;241m.\u001b[39mstderr)\n\u001b[1;32m 1214\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m blocking:\n\u001b[1;32m 1216\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mstatus_method\u001b[39m():\n",
|
1191 |
-
"\u001b[0;31mOSError\u001b[0m: remote: Scanning LFS files for validity, may be slow... \nremote: LFS file scan complete. \nremote: ----------------------------------------------------------\u001b[0;31m \nremote: Sorry, your push was rejected during YAML metadata verification: \nremote: - Error: \"language[0]\" with value \"zh-tw\" is not valid. It must be an ISO 639-1, 639-2 or 639-3 code (two/three letters), or a special value like \"code\", \"multilingual\". If you want to use BCP-47 identifiers, you can specify them in language_bcp47.\u001b[0;32m \nremote: ---------------------------------------------------------- \nremote: Please find the documentation at: \nremote: https://huggingface.co/docs/hub/model-cards#model-card-metadata\u001b[0;0m \nremote: ---------------------------------------------------------- \nTo https://huggingface.co/kimbochen/whisper-small-zh-tw\n ! [remote rejected] main -> main (pre-receive hook declined)\nerror: failed to push some refs to 'https://huggingface.co/kimbochen/whisper-small-zh-tw'\n"
|
1192 |
-
]
|
1193 |
}
|
1194 |
],
|
1195 |
"source": [
|
|
|
1141 |
},
|
1142 |
{
|
1143 |
"cell_type": "code",
|
1144 |
+
"execution_count": null,
|
1145 |
"id": "95737cda-c5dd-4887-a4d0-dfcb0d61d977",
|
1146 |
"metadata": {},
|
1147 |
"outputs": [
|
|
|
1156 |
"tokenizer config file saved in ./tokenizer_config.json\n",
|
1157 |
"Special tokens file saved in ./special_tokens_map.json\n",
|
1158 |
"added tokens file saved in ./added_tokens.json\n",
|
1159 |
+
"Several commits (13) will be pushed upstream.\n",
|
1160 |
"The progress bars may be unreliable.\n",
|
1161 |
"remote: Scanning LFS files for validity, may be slow... \n",
|
1162 |
"remote: LFS file scan complete. \n",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1163 |
"To https://huggingface.co/kimbochen/whisper-small-zh-tw\n",
|
1164 |
+
" 214645d..6f314a7 main -> main\n",
|
|
|
1165 |
"\n"
|
1166 |
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1167 |
}
|
1168 |
],
|
1169 |
"source": [
|