Speech recognition and synthesis
The state of the art of speech recognition by computers is reviewed. The human vocal and auditory mechanisms are described, and digitizing of voice signals is introduced. Techniques used in vocoders to achieve a bit rate 5 k bits/sec are summarized. Channel, linear prediction, and farmant vocoders are discussed, as are standardization and integration of vocoders.
NASA STI/Recon Technical Report N
- Pub Date:
- Speech Recognition;
- Voice Data Processing;
- Analog To Digital Converters;
- Communications and Radar