Definition
Process of creating statistical representations of speech sounds for recognition systems.
Detailed Explanation
Creates mathematical models that map acoustic features to phonetic units, typically using Hidden Markov Models or deep neural networks. Accounts for variations in pronunciation, speaking rate, and acoustic conditions.
Use Cases
Speech recognition systems, voice assistants, automated transcription services, voice-controlled devices