Commit
•
41fe0a6
1
Parent(s):
43087ad
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- mozilla-foundation/common_voice_11_0
|
| 5 |
+
language:
|
| 6 |
+
- yue
|
| 7 |
+
metrics:
|
| 8 |
+
- cer
|
| 9 |
+
library_name: transformers
|
| 10 |
+
pipeline_tag: automatic-speech-recognition
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<p align="left">
|
| 14 |
+
🤗 <a href="https://huggingface.co/Oblivion208" target="_blank">HF Repo</a> •🐱 <a href="https://github.com/fengredrum/finetune-whisper-lora" target="_blank">Github Repo</a>
|
| 15 |
+
</p>
|
| 16 |
+
|
| 17 |
+
## Approximate Performance Evaluation
|
| 18 |
+
|
| 19 |
+
The following models are all trained and evaluated on a single RTX 3090 GPU.
|
| 20 |
+
|
| 21 |
+
### Cantonese Test Results Comparison
|
| 22 |
+
|
| 23 |
+
#### MDCC
|
| 24 |
+
|
| 25 |
+
| Model name | Parameters | Finetune Steps | Time Spend | Training Loss | Validation Loss | CER % | Finetuned Model |
|
| 26 |
+
| ------------------------------- | ---------- | -------------- | ---------- | ------------- | --------------- | ----- | ------------------------------------------------------------------------------------------------------------------------ |
|
| 27 |
+
| whisper-tiny-cantonese | 39 M | 3200 | 4h 34m | 0.0485 | 0.771 | 11.10 | [Link](https://huggingface.co/Oblivion208/whisper-tiny-cantonese "Oblivion208/whisper-tiny-cantonese") |
|
| 28 |
+
| whisper-base-cantonese | 74 M | 7200 | 13h 32m | 0.0186 | 0.477 | 7.66 | [Link](https://huggingface.co/Oblivion208/whisper-base-cantonese "Oblivion208/whisper-base-cantonese") |
|
| 29 |
+
| whisper-small-cantonese | 244 M | 3600 | 6h 38m | 0.0266 | 0.137 | 6.16 | [Link](https://huggingface.co/Oblivion208/whisper-small-cantonese "Oblivion208/whisper-small-cantonese") |
|
| 30 |
+
| whisper-small-lora-cantonese | 3.5 M | 8000 | 21h 27m | 0.0687 | 0.382 | 7.40 | [Link](https://huggingface.co/Oblivion208/whisper-small-lora-cantonese "Oblivion208/whisper-small-lora-cantonese") |
|
| 31 |
+
| whisper-large-v2-lora-cantonese | 15 M | 10000 | 33h 40m | 0.0046 | 0.277 | 3.77 | [Link](https://huggingface.co/Oblivion208/whisper-large-v2-lora-cantonese "Oblivion208/whisper-large-v2-lora-cantonese") |
|
| 32 |
+
|
| 33 |
+
#### Common Voice Corpus 11.0
|
| 34 |
+
|
| 35 |
+
| Model name | Original CER % | w/o Finetune CER % | Jointly Finetune CER % |
|
| 36 |
+
| ------------------------------- | -------------- | ------------------ | ---------------------- |
|
| 37 |
+
| whisper-tiny-cantonese | 124.03 | 66.85 | 35.87 |
|
| 38 |
+
| whisper-base-cantonese | 78.24 | 61.42 | 16.73 |
|
| 39 |
+
| whisper-small-cantonese | 52.83 | 31.23 | / |
|
| 40 |
+
| whisper-small-lora-cantonese | 37.53 | 19.38 | 14.73 |
|
| 41 |
+
| whisper-large-v2-lora-cantonese | 37.53 | 19.38 | 9.63 |
|