scart97/thunder-speech
A Hackable speech recognition library.
Provides end-to-end ASR models with straightforward fine-tuning for new languages and seamless export to TorchScript for production inference. Supports loading pretrained checkpoints from both local registry and Hugging Face Hub, with decoupled audio processing pipelines that can be independently scripted and deployed. Emphasizes developer ergonomics through a PyTorch-native API that exposes raw model outputs for custom decoding while maintaining sensible defaults.
No commits in the last 6 months. Available on PyPI.
Stars
25
Forks
3
Language
Python
License
MIT
Category
Last pushed
Oct 16, 2024
Monthly downloads
795
Commits (30d)
0
Dependencies
12
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/scart97/thunder-speech"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
githubharald/CTCWordBeamSearch
Connectionist Temporal Classification (CTC) decoder with dictionary and language model.
githubharald/CTCDecoder
Connectionist Temporal Classification (CTC) decoding algorithms: best path, beam search, lexicon...
nl8590687/ASRT_SpeechRecognition
A Deep-Learning-Based Chinese Speech Recognition System 基于深度学习的中文语音识别系统
athena-team/athena
an open-source implementation of sequence-to-sequence based speech processing engine
hirofumi0810/tensorflow_end2end_speech_recognition
End-to-End speech recognition implementation base on TensorFlow (CTC, Attention, and MTL training)