Downloads · 30 days
41
9% of all-time downloads
NUTN-KWS/Whisper-Taiwanese-Hakka-model-v0.2.6
Whisper-Taiwanese-Hakka-model-v0.2.6 is a automatic speech recognition model from NUTN-KWS. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as cc-by-nc-4.0.
Downloads · 30 days
41
9% of all-time downloads
All-time downloads
473
Public
Parameters
809M
3.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors3.2 GB · 100%
From the Hugging Face model README
[ 英文 README_EN.md ]
這個模型是由國立臺南大學執行國科會產學合作計畫,使用 openai/whisper-large-v3-turbo 微調的版本,並執行國科會TAIDE台英語家庭先導計畫,與真平出版社合作,使用中小學教材內容、優聲學工具生成客語語音及學生學習資料進行模型微調,用於真平教材客語辨識。並與國研院國網中心合作,運用國網中心算力以及TAIDE模型,共同建構中小學台英客語AI學習模型。
示範網址: https://kws.oaselab.org/taigitong/hakka
openai/whisper-large-v3-turbopip install torch torchvision torchaudio transformers
from transformers import pipeline
pipe = pipeline("automatic-speech-recognition", model="./model/whisper-taiwanese-hakka", device=0)
result = pipe("audio.wav", generate_kwargs={"language": "zh", "task": "transcribe"})
print(result["text"])
@misc{taiwanesehakkawhisperasr2026,
title={Taiwanese Hakka ASR},
author={KWS Center, National University of Tainan, Taiwan},
year={2026},
url={https://huggingface.co/NUTN-KWS/Whisper-Taiwanese-Hakka-model-v0.2.6}
}