Speech Analytics

Definition - What does Speech Analytics mean?

Speech analytics is the process of extracting relevant and vital content from a recorded audio file. It has the capability to automatically identify, categorize and cross-reference important information regarding what is being said or the actual substance or meaning of the speech, not just individual words. In order to make this possible, speech analytics makes use of several types of software applications like automatic speech recognition and audio mining tools.

Techopedia explains Speech Analytics

There are three major approaches in speech analytics: speech-to-text, direct phrase recognition, and phonetics.

Speech-to-text: Uses bi-grams or tri-grams as basic units in analyzing speech and needs to have it matched to thousands of words. The outcome is a flow of words making it easier to work with and is also accurate.
Direct phase recognition: Directly analyzes speech by searching for predefined phrases rather than converting it to text or phonemes. Although this is the longest method, it is also the most accurate since no information is lost while converting the data.
Phonetic: The quickest way in processing since the basic unit used is a phoneme. Given that there are only a few phonemes known in most languages, a long list of these is used which the software cross-references to target phonemes to the closest one in the list.