CMUSphinx is a speaker-independent large vocabulary continuous speech recognizer released under BSD style license. It is also a collection of open source tools and resources that allows researchers and developers to build speech recognition systems.
- speech recognition
- audio transcription
- captions alignment
PocketSphinx is the best you can get as alternative for online speech-to-text solutions. I use it in developed command and control desktop applications (with US/UK English interface). Very fast, especially with small dictionaries and keywords search. Downsides: A little expensive in resources (CPU and memory) if running in NGram search mode. Only few supported languages (others require creating own language models which is not a trivial task even for experienced developers, i.e. I simply give up when designing a model for Polish language).
Great tool for speech recognition with great support from contributor.
Kudos to the CMU Sphinx team. No other offline speech recognition library is more easy to use, more customizable and more accurate.