PyThaiNLP/pythaiasr

Python Thai Automatic Speech Recognition

67
/ 100
Established

Supports both file-based and real-time streaming transcription via microphone input, with configurable audio chunk processing. Built on Wav2Vec2 and Whisper architectures from Hugging Face, offering multiple pre-trained model variants with optional language model integration for improved accuracy. Integrates seamlessly with PyTorch and provides flexible device selection for CPU or GPU inference.

Available on PyPI.

Maintenance 10 / 25
Adoption 16 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

77

Forks

13

Language

Python

License

Apache-2.0

Last pushed

Jan 31, 2026

Monthly downloads

856

Commits (30d)

0

Dependencies

6

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/PyThaiNLP/pythaiasr"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.