Instructions to use chienhsinlin/helioslm-asr-tw-v5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use chienhsinlin/helioslm-asr-tw-v5 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="chienhsinlin/helioslm-asr-tw-v5")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("chienhsinlin/helioslm-asr-tw-v5") model = AutoModelForSpeechSeq2Seq.from_pretrained("chienhsinlin/helioslm-asr-tw-v5", device_map="auto") - Notebooks
- Google Colab
- Kaggle
羲和 HeliosLM ASR TW v5
羲和(HeliosLM)ASR TW v5 是以 OpenAI Whisper large-v3 為基礎、針對台灣華語(繁體中文輸出)與台語(漢字輸出)微調的語音辨識模型。它是羲和對話模型 helioslm-chat-e4b 的語音搭配模型。
HeliosLM ASR TW v5 is a speech recognition model fine-tuned from openai/whisper-large-v3 for Taiwanese Mandarin (Traditional Chinese output) and Taiwanese Hokkien (Han-character output).
- 參數量 Parameters: about 1.55B (Whisper large-v3 architecture)
- 授權 License: Apache-2.0 (same as the base model)
訓練資料 Training data
All public data, no private recordings:
- Mozilla Common Voice 27.0 zh-TW and nan-tw (CC0-1.0), test sentences and test speakers excluded
- Google FLEURS cmn_hans_cn train/dev (CC-BY-4.0), transcripts converted to Traditional Chinese with OpenCC
- AISHELL-1 train subset (Apache-2.0)
1 epoch, learning rate 7e-6, 4,776 steps; then the fine-tuned weights were linearly merged with the base model (0.85 × fine-tuned + 0.15 × whisper-large-v3).
評測 Evaluation (2026-11-05, self-run)
Character error rate (CER, %, lower is better), greedy decoding, same clips and same run for both models. Scores are self-reported.
| Test set | Clips | HeliosLM ASR TW v5 | whisper-large-v3 (base) |
|---|---|---|---|
| Common Voice 27.0 zh-TW test | 998 | 5.10 | 15.25 |
| FLEURS cmn test (Traditional Chinese refs) | 945 | 7.33 | 7.54 |
| Common Voice 27.0 nan-tw test, Han characters | 1,000 | 38.15 | 231.13 |
| Common Voice 27.0 nan-tw test, raw transcripts (incl. romanization) | 1,000 | 86.29 | 121.57 |
Taiwanese Hokkien is still weak: the model writes Han characters only and does not output Tâi-lô romanization.
使用 Usage
from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="chienhsinlin/helioslm-asr-tw-v5", chunk_length_s=30)
print(asr("audio.wav", generate_kwargs={"language": "zh", "task": "transcribe"})["text"])
模型可能辨識錯誤,重要內容請自行確認。The model can make mistakes; verify important transcripts.
- Downloads last month
- 38
Model tree for chienhsinlin/helioslm-asr-tw-v5
Base model
openai/whisper-large-v3