List of speech recognition software

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Open source acoustic models and speech corpus[edit]

The following list presents notable speech recognition software engines with a brief synopsis of characteristics.

Application name Description Website Open Source License Operating System Note
CMU Sphinx HMM CMU: Sourceforge Yes Linux Java
HTK HMM HTK Web Site Yes HTK Specific License Multi-platform C. Inactive since April 2009.
Julius HMM trigrams Julius Home page Yes BSD-like Multi-platform C
Kaldi Deep neural net. Kaldi Sourceforge Yes Apache Multi-platform C++
iATROS LDA (Latent Dirichlet) iATROS Yes
RWTH ASR RWTH Aachen University RWTH ASR No RWTH ASR License Linux, Mac OS X Non-commercial use only

The following lists open-source applications that provide convenient user interfaces for the above.

Application name Description Website Open Source License Operating System Note
Simon Supports Sphinx, HTK, Julius Simon Yes GPLv2 Multi-platform


Application name Description Website Open Source License Price Note
Dragon Dictate Mac OS No Proprietary
MacSpeech Dictate Medical Medical dictation product
Macspeech Dictate Legal Legal-focused dictation
MacSpeech Scribe Transcription from recorded text
iListen PowerPC Macintosh
Speakable items Included with Mac OS
ViaVoice IBM Product. Support ended 2007.
Voice Navigator Original GUI voice control (1989)
Power Secretary[1]
Vestec Inc. ASR, NLU, TTS, VSLIC [2]

Mobile devices and smartphones[edit]

Many cell-phone handsets have basic dial-by-voice features built in. Smartphones such as iPhones and Blackberrys also support this. A number of third-party apps have implemented natural-language speech-recognition support, including: