huggingface/distil-whisper
Distilled variant of Whisper for speech recognition. 6x faster, 50% smaller, within 1% word error rate.
Knowledge distillation from OpenAI's Whisper model enables this variant to achieve competitive accuracy on English speech through parameter reduction and architectural optimization. Fully compatible with Hugging Face Transformers' `AutoModelForSpeechSeq2Seq` API, it supports both short-form (<30s) and long-form transcription via sequential windowing, with optional features like beam search, timestamp generation, and prompt-based conditioning through the standard generation interface.
4,056 stars. No commits in the last 6 months.
Stars
4,056
Forks
349
Language
Python
License
MIT
Category
Last pushed
Jan 08, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/huggingface/distil-whisper"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
YuanGongND/whisper-at
Code and Pretrained Models for Interspeech 2023 Paper "Whisper-AT: Noise-Robust Automatic Speech...
adi-gov-tw/Taiwan-Tongues-ASR-CE
Taiwan Tongues ASR CE 是一個開源語音辨識（Automatic Speech Recognition, ASR）模型專案，專為台灣多元語言環境設計。 本模型支援...
KevKibe/African-Whisper
🚀 Framework for seamless fine-tuning of Whisper model on a multi-lingual dataset and deployment to prod.
sandy1990418/ChineseTaiwaneseWhisper
This repository focuses on leveraging OpenAI's Whisper model for speech recognition in Chinese...
ga642381/Taiwanese-Whisper
fine-tune Whipser model for Taiwanese speech recognition