Speech recognition

Refers to the conversion of spoken words and phrases into text. Speech recognition software (also known as automatic speech recognition or computer speech recognition) converts spoken words to machine-readable input.

This method can be used to prepare structured documents, as well as for speech-to-text processing (e.g. when using a word processor or email application). Searches, queries, commands and form filling can sometimes be faster to perform by voice than by using a keyboard. Alternatively, the resulting text can be used as an input for further linguistic processing.

The W3C has released a Speech Recognition Grammar Specification (SRGS), which is a set of word patterns that tells the speech recognition system what to expect the speaker to say.

Audio visual speech recognition (AVSR) uses image processing capabilities in lip reading to aid speech recognition systems. This is particularly useful when the exact word is unclear from the user’s speech.

Related methods include: Searching and querying.

tool: Praat

A software application that may be used to analyse, synthesise, and manipulate speech in phonetics, and create visual representations of the audio recording.
Methods relating to this toolCategory
Music recognitionData capture
Sound analysisData analysis
Sound editingData structuring and enhancement
Sound editingPractice-led research
Speech recognitionData capture
Lifecycle stage:
Alternate tool(s):

Speech Analyzer

Syndicate content