About this document
Speech Signal Processing Overview by duy quang nguyễn is a document available to read on EtoBox.
Speech is converted from analog to digital using an analog-to-digital converter that samples the sound multiple times per second. Speech recognition systems initially used time domain features but now use frequency domain analysis by splitting the waveform into short frames, windowing them, and performing FFT or mel cepstral analysis to distinguish phonetic types while being invariant to other factors. The frequency domain allows distinguishing sounds like "beat" versus "bat" based on their spectragrams.
- Author
- duy quang nguyễn
- Language
- EN