Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ language:
|
|
| 11 |
|
| 12 |
# PhoASR-whisper-small: Vietnamese Automatic Speech Recognition - A Revisit
|
| 13 |
|
| 14 |
-
> We introduce `PhoASR-whisper-small`, a Vietnamese automatic speech recognition model obtained by fine-tuning the multilingual `whisper-small` on a 3000-hour dataset (`PhoASR-3000h`). The model directly produces clean transcripts with reliable timestamps, punctuation and capitalization, eliminating the need for separate post-processing tools. Please **cite** our [PhoASR paper](https://aclanthology.org/2026.findings-eacl.345/) whenever `PhoASR-whisper-small` is used to produce published results or is incorporated into other software
|
| 15 |
|
| 16 |
@inproceedings{phoasr,
|
| 17 |
title = {{Vietnamese Automatic Speech Recognition: A Revisit}},
|
|
@@ -24,7 +24,8 @@ language:
|
|
| 24 |
## Model Inference with `transformers`
|
| 25 |
|
| 26 |
```python
|
| 27 |
-
#
|
|
|
|
| 28 |
import torch
|
| 29 |
from transformers import pipeline
|
| 30 |
|
|
@@ -45,8 +46,10 @@ print(f"Full Text: {result['text']}")
|
|
| 45 |
print(result['chunks'])
|
| 46 |
```
|
| 47 |
|
|
|
|
|
|
|
| 48 |
## License/Terms of Use
|
| 49 |
-
This model is released under the BSD 3-Clause Clear license and the Qualcomm responsible AI license: [https://www.qualcomm.com/site/responsible-ai-license](https://www.qualcomm.com/site/responsible-ai-license)
|
| 50 |
|
| 51 |
## Uses
|
| 52 |
The model is intended for research and educational purposes.
|
|
|
|
| 11 |
|
| 12 |
# PhoASR-whisper-small: Vietnamese Automatic Speech Recognition - A Revisit
|
| 13 |
|
| 14 |
+
> We introduce `PhoASR-whisper-small`, a Vietnamese automatic speech recognition model obtained by fine-tuning the multilingual `whisper-small` on a 3000-hour dataset (`PhoASR-3000h`). The model directly produces clean transcripts with reliable timestamps, punctuation and capitalization, eliminating the need for separate post-processing tools. Please **cite** our [PhoASR paper](https://aclanthology.org/2026.findings-eacl.345/) whenever `PhoASR-whisper-small` is used to produce published results or is incorporated into other software.
|
| 15 |
|
| 16 |
@inproceedings{phoasr,
|
| 17 |
title = {{Vietnamese Automatic Speech Recognition: A Revisit}},
|
|
|
|
| 24 |
## Model Inference with `transformers`
|
| 25 |
|
| 26 |
```python
|
| 27 |
+
# INSTALLATION: pip install transformers==4.48.0
|
| 28 |
+
|
| 29 |
import torch
|
| 30 |
from transformers import pipeline
|
| 31 |
|
|
|
|
| 46 |
print(result['chunks'])
|
| 47 |
```
|
| 48 |
|
| 49 |
+
For more low-level control, please see [inference.py](https://huggingface.co/Qualcomm-AI-Research/PhoASR-whisper-small/blob/main/inference.py).
|
| 50 |
+
|
| 51 |
## License/Terms of Use
|
| 52 |
+
This model is released under the BSD 3-Clause Clear license and the Qualcomm responsible AI license: [https://www.qualcomm.com/site/responsible-ai-license](https://www.qualcomm.com/site/responsible-ai-license).
|
| 53 |
|
| 54 |
## Uses
|
| 55 |
The model is intended for research and educational purposes.
|