Downloads · 30 days
2
5% of all-time downloads
reworkwhisper-large-v1
winhsss/reworkwhisper-large-v1
reworkwhisper-large-v1 is a machine learning model from winhsss. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
LoRA fine-tune của vinai/PhoWhisper-small trên bộ meeting-ASR tiếng Việt dataset-dot1.
All-time downloads
37
Public
Parameters
1.5B
3.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.1 GB · 100%
At a glance
- Library
- peft
- Model type
- whisper
- Access
- Public
- Created
- Jul 23, 2026
- Updated
- Jul 23, 2026
- SHA
- 0f848f4e
Base models
Model card
From the Hugging Face model README
PhoWhisper-large + LoRA (dataset-dot1)
LoRA fine-tune của vinai/PhoWhisper-small trên bộ meeting-ASR tiếng Việt dataset-dot1.
Kết quả (test set 500 seg, greedy, CER chuẩn hoá)
| model | CER% | WER% | RTF |
|---|---|---|---|
| base | 36.22 | 56.93 | 0.026 |
| + LoRA | 31.38 | 47.41 | 0.089 |
LoRA r=32, alpha=64, epochs=3, lr=0.0001. Loader: HF PEFT (transformers + bitsandbytes 4-bit).
⚠️ Data synthetic (viXTTS) — số liệu chỉ để so tương đối, không phải hiệu năng trên giọng người thật.
Dùng
from transformers import WhisperForConditionalGeneration, WhisperProcessor
from peft import PeftModel
base = WhisperForConditionalGeneration.from_pretrained("vinai/PhoWhisper-small")
model = PeftModel.from_pretrained(base, "winhsss/viet-speech-small-v1")
processor = WhisperProcessor.from_pretrained("winhsss/viet-speech-small-v1")
- Library
- peft
- Type
- whisper
- Languages
- vi
- Created
- Jul 23, 2026
- Updated
- Jul 23, 2026