biodatlab/thonburian-whisper
Thonburian Whisper: Open models for fine-tuned Whisper in Thai. Try our demo on Huggingface space:
Fine-tuned on a composite dataset spanning CommonVoice, Gowajee, Thai Elderly Speech, and dialect corpora, with variants optimized for noise robustness and domain-specific applications (financial, medical). Integrates directly with Hugging Face transformers pipeline for inference, and offers both full-scale and distilled model checkpoints (166M-1.5B parameters) with measurable WER benchmarks and inference speed trade-offs optimized for different deployment scenarios.
186 stars. No commits in the last 6 months.
Stars
186
Forks
20
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jul 29, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/biodatlab/thonburian-whisper"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
Arkapravo-Ghosh/speech-to-text
Speech to Text Transcription using OpenAI Whisper v3 and FastAPI
haiodo/oaitt
An OpenAI compatible transcriber using transformers and whisperx.
purvanshjoshi/IndiVoice-DeepASR
Deep Learning framework for Indian-accented Speech-to-Text using Whisper and LoRA. Includes...
boned-fruitwood759/whisperx-asr-with-fastapi
🎤 Enable real-time speech recognition with WhisperX using FastAPI for efficient, scalable audio...
tomdewildt/whisper-experiment
Experiments using the Whisper model from Open AI