VoxSigma^® Speech Recognition Software Suite

The VoxSigma software suite offers large vocabulary speech-to-text capabilities in many languages. It has been designed for professional users needing to transcribe large quantities of audio and video documents such as broadcast data, either in batch mode or in real-time. Versions can also be used to transcribe telephone data.

The full speech-to-text conversion is done in three steps. The software first identifies the audio segments containing speech, then it identifies the language being spoken if it's not known, and eventually it converts the speech segments to text. It includes adaptive features allowing the transcription of noisy speech like speech with background music.

The speech-to-text conversion result is a fully annotated XML document including labels for speech and non-speech segments, speaker labels, words with time codes and confidence scores. This XML file may be directly indexed by a search engine, or it can be converted into plain text with punctuation.

Platforms	Linux x86 and x86_64 (OpenSuse, Debian, Fedora, CentOS, Ubuntu, SuSE, Red Hat, ...)
API	command line tools, C++ library
Audio	studio (e.g. broadcast) and telephone bandwidths
Key functions	audio segmentation, speaker segmentation, language identification, spoken word transcription (speech-to-text)
Operating modes	batch, real-time, single or multi-threaded
Ouputs	XML with speaker diarization, language identification tags, word transcription, punctuation, confidence measures, numeral entities and other specific entities
Supported Languages	Arabic, Dutch, English (US, UK), Finnish, French, German, Greek, Italian, Lithuanian, Mandarin, Polish, Portuguese, Romanian, Russian, Spanish, Turkish (other language options are under development, contact us for more information)

To get more information about the VoxSigma software suite or to get a price quote you may fill out our online request form.