Variable time delay and bit rate speech communication on ATM networks
Yang Zhen, Huang Xiwei, Bi Houjie
Abstract
Yang Zhen, Huang Xiwei, Bi Houjie
Abstract
There are some new demands on speech processing when voice communication is carried out over ATM networks. This paper discusses some problems of speech coding and transmission on ATM networks. We propose two schemes for speech processing, which have variable transmission delay and bit rate respectively, to reach at the goals: small assembly delay, high transmission efficiency and toll speech quality. We transmit different speech cells in different time intervals and define a new "distribution entropy" to classify different input signals. We then design a new speech codec in the second scheme whose bit rate is perceptual importance based. It has better speech quality than that of the G.727 system.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
There are some new demands on speech processing when voice communication is carried out over ATM networks. This paper discusses some problems of speech coding and transmission on ATM networks. We propose two schemes for speech processing, which have variable transmission delay and bit rate respectively, to reach at the goals: small assembly delay, high transmission efficiency and toll speech quality. We transmit different speech cells in different time intervals and define a new "distribution entropy" to classify different input signals. We then design a new speech codec in the second scheme whose bit rate is perceptual importance based. It has better speech quality than that of the G.727 system.
Key concepts: Computer science, Codec2, Speech coding, PSQM, Full Rate, Linear predictive coding, Harmonic Vector Excitation Coding, Speech recognition