Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Advances in Non-Linear Modeling for Speech ProcessingAM-FM: Modulation and Demodulation Techniques

Advances in Non-Linear Modeling for Speech Processing: AM-FM: Modulation and Demodulation Techniques [Analysis of speech signals is usually carried out using STFT. The most successful features currently being used in both speech recognition and speaker recognition systems are cepstral features. The cepstral features in one way or another are based on the source-filter model of speech production. However, it is well known that a significant part of the acoustic information cannot be modeled by the linear source-filter model. The source-filter model assumes that the sound source for the voiced speech is localized in the larynx and the vocal tract acts as a convolution filter for the emitted sound. Examples of phenomena not well-captured by the source-filter model include unstable airflow, turbulence and nonlinearities arising from oscillators with time-varying masses.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Advances in Non-Linear Modeling for Speech ProcessingAM-FM: Modulation and Demodulation Techniques

Download PDF
14 pages

Loading next page...
 
/lp/springer-journals/advances-in-non-linear-modeling-for-speech-processing-am-fm-modulation-7PlCH74V3h
Publisher
Springer US
Copyright
© The Author(s) 2012
ISBN
978-1-4614-1504-6
Pages
61 –75
DOI
10.1007/978-1-4614-1505-3_5
Publisher site
See Chapter on Publisher Site

Abstract

[Analysis of speech signals is usually carried out using STFT. The most successful features currently being used in both speech recognition and speaker recognition systems are cepstral features. The cepstral features in one way or another are based on the source-filter model of speech production. However, it is well known that a significant part of the acoustic information cannot be modeled by the linear source-filter model. The source-filter model assumes that the sound source for the voiced speech is localized in the larynx and the vocal tract acts as a convolution filter for the emitted sound. Examples of phenomena not well-captured by the source-filter model include unstable airflow, turbulence and nonlinearities arising from oscillators with time-varying masses.]

Published: Feb 21, 2012

Keywords: Amplitude Modulation; Speech Signal; Instantaneous Frequency; Vocal Tract; Speaker Identification

There are no references for this article.