Speech Recognition

Speech recognition is a mapping from a digitally-encoded acoustic signal to string of words. This can be viewed as a diagnosis problem. Speech understanding also includes:

what speech sounds were uttered
choosing words based on intention
choosing meaning of words based on intention
natural language

The sounds in human language are limited to 40 or 50 sounds, which are called phones. There are 49 phones for English:

Speech recognition is the problem of identifying phones based on features of the acoustic signal, such as frequency or amplitude.