A shaped trellis coded quantizer for sampled speech /
adaptive scheme in order to provide excellent quality speech.
| Main Author: | |
|---|---|
| Format: | Thesis eBook |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1994.
|
| Subjects: | |
| Online Access: | Link to OAKTrust copy |
| Summary: | adaptive scheme in order to provide excellent quality speech. an overall delay of 5 ms for this coder. Simulation results called the "S-TCQ speech coder" uses an efficient backward coder compares favorably with other existing 16 kbit/sec coding schemes for Gauss-Markov sources. This powerful delay a backward block adaptive scheme is used. Simulation developed. This technique incorporates frequency-weighting Due to recent developments in digital communication encoded speech has little background noise and is natural for more efficient speech coders. In this work, a new 16 generalized speech coder using noz'se feedback codz'ng is good model for sampled speech. The performance of this has been studied on a Gauss-Markov source which serves as a In this work, as a first step towards the development of a kbit/sec low delay speech coder is developed. This coder, namely, shaped trellis coded quan.tizatz'on was developed obtained for standard test speech files. Further, a of noise. Smaller block length is used in order to achieve order to meet with the low delay requirement, there is a need predictor and a variance estimator are used to exploit the quantization scheme has then been applied for the coding of recently by Larola et al. [1]. They have demonstrated that redundancies in speech waveforms. In order to provide low required quality, but are unable to provide low delays. In results for this coder show excellent performance. This sampled speech. In addition, a high order short-term scheme has been found to be better than most other source show a noticeable improvement in the perceptual quality. The sounding. speech at low bit rates. Most speech algorithms based on speech coder, the shaped trellis coded quantization scheme speech coders. SNR values in the range 16.0-19.5 dB are technology there is increasing need to achieve high quality The inherent quantization scheme in the S-TCQ speech coder, the S-TCQ quantizer achieves high granular gains (due to trellis coding) and high shaping gains (due to SVQ shaping). VSELP and other similar systems are able to achieve the |
|---|---|
| Item Description: | "Major subject: Electrical Engineering". Vita. |
| Physical Description: | x, 59 leaves : illustrations ; 28 cm. Also available online. |
| Bibliography: | Includes bibliographical references. |