Open-source speech coding-speex

Speex is an open-source free speech (audio) CODEC library. Official Website:

Http://www.speex.org/

It is part of the GNU project, but you don't have to worry about using it in your application, because it is not based on GNU license management, but on revisedbsd license.

CELP (Code-Excited Linear Prediction) is used in speex encoding and decoding. translating it into Chinese is code-driven linear prediction. Speex encoding can be used to obtain the speech data with a bit rate of 2-kbps. Is the abbreviation of code excited linear prediction. CELP is the most successful Speech Encoding Algorithm in the past 10 years. The CELP Speech Encoding Algorithm uses linear prediction to extract audio channel parameters. It uses a code book containing many typical excitation vectors as the excitation parameter and searches for an optimal excitation vector in this code book for each encoding, the encoded value of this excitation vector is the serial number in the code book of this sequence. For more information about the CELP algorithm, see:

Http://baike.baidu.com/view/2171835.htm

Http://en.wikipedia.org/wiki/Code-excited_linear_prediction

ECHO is a difficult problem in speech coding, especially in audio and video conferences. However, if you use speex, this problem can be well solved. In 2008, speex 1.2rc1 was released in July 23, which effectively solved this problem.

Speex has the characteristics that are not available in other speech (audio) coding and decoding methods. It encodes three sampling rates, low frequency 8 kHz, 16 kHz, and 32 kHz in the same speech stream. There is also a bit of intensity stereo encoding, the literal meaning is the intensity of three-dimensional encoding, which is not very clear.

The official website provides source code download. This library is compiled by C ++, so if your program is written in C ++, you can directly embed the source code of the speex library into your code, which is easy to debug.

.