jonatasgrosman/huggingsound
Python library for speech recognition using CTC and transformer models from Hugging Face.
Not currently ranked — collecting fresh signals.
star history
A speech processing toolkit built on Hugging Face infrastructure that provides ready-to-use interfaces for automatic speech recognition tasks. It supports transcription with CTC models such as wav2vec2, character-level timestamps and probabilities, and language model decoding for improved accuracy. The library also includes speaker diarization, speech enhancement, and finetuning capabilities.
Frequently asked
- What is jonatasgrosman/huggingsound?
- Python library for speech recognition using CTC and transformer models from Hugging Face.
- Is huggingsound open source?
- Yes — jonatasgrosman/huggingsound is open source, released under the MIT license.
- What language is huggingsound written in?
- jonatasgrosman/huggingsound is primarily written in Python.
- How popular is huggingsound?
- jonatasgrosman/huggingsound has 470 stars on GitHub.
- Where can I find huggingsound?
- jonatasgrosman/huggingsound is on GitHub at https://github.com/jonatasgrosman/huggingsound.