| 摘要 |
The method consists in segmenting the speech on transmission into frames of fixed length so as to transmit them independently of one another and in reconstructing the speech on reception with a synthesising filter (29). Each frame comprises a first field (field#/ 1) made up of binary words with identical structures for coding the position and the amplitude of the pulses for exciting the synthesising filter, a second field (field#/ 2) made up of at least one binary word for coding the amplitude of the reference pulse associated with the frame and a third field (field#/ 3) made up of binary words for coding the coefficients of the synthesising filter. Only the codes representing the reference amplitude and the coefficients of the synthesising filter making up the second and third field are accompanied by codes for auto-correction of errors, exploited on reception by detectors (23, 24) for correction of errors. Application: radiotelephones. <IMAGE> |
