gooofy/zamia-speech
Toolkit for building and training automatic speech recognition models from speech corpora.
Not currently ranked — collecting fresh signals.
star history
This repository provides scripts for constructing ASR audio models (Kaldi nnet3 chain, wav2letter++), language models (KenLM in ARPA format), and grapheme-to-phoneme models (sequitur g2p) from speech data sources like VoxForge. It processes speech and text corpora to train models that decode audio to text. Pre-trained models and a Docker-based STT demo service are also provided.
Frequently asked
- What is gooofy/zamia-speech?
- Toolkit for building and training automatic speech recognition models from speech corpora.
- Is zamia-speech open source?
- Yes — gooofy/zamia-speech is open source, released under the LGPL-3.0 license.
- What language is zamia-speech written in?
- gooofy/zamia-speech is primarily written in Python.
- How popular is zamia-speech?
- gooofy/zamia-speech has 448 stars on GitHub.
- Where can I find zamia-speech?
- gooofy/zamia-speech is on GitHub at https://github.com/gooofy/zamia-speech.