Automatic speech recognition
Automatic speech recognition(often incorrectly calledspeech recognition) is a computer technique that analyzes the human voice captured through a microphone and transcribes it into machine-readable text.
Speech recognition, as well as speech synthesis, speaker identification or speaker verification, are part of speech processing techniques. These techniques make it possible in particular to produce human-machine interfaces(HIM) where part of the interaction is done by voice: « voice interfaces».
Among the many applications, we can cite voice dictation applications on a computer where the difficulty lies in the size of the vocabulary and the length of the sentences, but also interactive voice server type telephone applications, where the difficulty is rather the need to recognize any voice under variable and often noisy acoustic conditions (mobile phones in public places).
Source: Wikipedia« Back to Glossary Index