Speech coding transmission equipment

Claim 2

A speech coding transmission equipment having a transmitting portion transmitting speech signals processed by time domain harmonic compression and coded, and a receiving portion decoding the transmitted coded signal, and expanding and reproducing the decoded signal to a bandwidth existing before compression of the signal, said transmitting portion comprising, a voiced/unvoiced determination means (10) for distinguishing whether the speech signal is in a voiced period or an unvoiced period, a pitch period extraction means (2) for extracting a pitch period of the speech signal during the voiced period, a time domain harmonic compression means (1,101) for obtaining and outputting a speech signal corresponding to one pitch period from a speech signal included in a plurality of pitch periods, a decimation means (7) for obtaining and outputting a speech signal corresponding to one sample period from a speech signal included in a plurality of sample periods during an unvoiced period, said receiving portion comprising, a time domain harmonic expansion means (6,106) for expanding, in a time domain, a speech signal corresponding to one pitch period into a plurality of pitch periods, and an interpolation means (8) for expanding, in the time domain, a speech signal corresponding to the one sample period into a plurality of sample periods.

Claim 3

A speech coding transmission equipment as set forth in claim 1, comprising, a coder means (3) receiving an output of a time domain harmonic compression means and encoding that output, a multiplexer (21) for multiplexing data output from the coder means (3) and a fixed pitch period data in a fixed frame, based on an output of a pitch period from a pitch period extraction means (2), a border data generation means (20) for detecting a border presence in the fixed length frame and for detecting a border position based on the output of the pitch period, and the frame length being made shorter than that of the shortest pitch period, the border presence data and the border position data being multiplexed as fixed pitch period data into the fixed frame length, by the multiplexer.

Claim 4

A speech coding transmission equipment as set forth in claim 2, wherein the coder means (3) comprises, a plurality of adaptive differential pulse code modulation means (300 - 30k), a portion or all of which are selected in response to the border presence data from the border data generation means(20), the equipment comprising, an optimum coder determination means(60) for determining an optimum coder in the plurality of adaptive differential pulse code modulation means (300 - 30k), a selection means (32) for selecting the output of the optimum coder (60), the multiplex means (21) further multiplexing the optimum coder data.

Claim 5

A speech coding transmission equipment as set forth in claim 3, wherein, when the border of the pitch period is not present, all of the adaptive differential pulse code modulation means (300-30k) are connected to the time domain harmonic compression means (1), and when the border of the pitch period is present, a portion of the adaptive differential pulse code modulation means (300-30k) is connected to the time domain harmonic compression means (1), and excess bits caused by reducing allocation of the adaptive differential pulse code modulation means (300 - 30k) is allotted to the border position data of the pitch period.

Claim 6

A speech coding transmission equipment as set forth in claim 3, wherein, the optimum coder determination means (60) comprises a quantized error power calculation means (61) for calculating a quantized error power of each adaptive differential pulse code modulation means (300 - 30k) and an optimum quantization determination means (62) for determining the optimum coder (60) as an optimum quantization means (62).

Claim 7

A speech coding transmission equipment having a transmitting portion transmitting speech signal processed by a time domain harmonic compression and coded, and a receiving portion decoding the transmitted coded signal, and expanding and reproducing the decoded signal to a bandwidth existing before compression of the signal, said transmitting portion comprising, a time domain harmonic compression and expansion control means (102) distinguishing whether the speech signal is in the period of voiced speech or period of unvoiced speech, whereby, if the speech signal is the period of voiced speech a pitch period of the speech signal is extracted and output, and if the speech signal is in the period of unvoiced speech, a sample period signal by which the speech signal is sampled is output, and a time domain harmonic compression means (101) receiving the output of the pitch period from the time domain harmonic compression and expansion control means (102), obtaining the speech signal corresponding to one pitch period from the speech signal included in a plurality of pitch periods, and obtaining and outputting the speech signal corresponding to one sample period from the speech signal included in a plurality of sample periods receiving the output of the sample period from the time domain harmonic compression and expansion control means (102), and said receiving portion comprising, a time domain harmonic expansion means (106) receiving the output of the pitch period from the time domain harmonic compression and expansion control means (102), expanding the speech signal corresponding to one pitch period to a plurality of pitch periods, receiving the output of the sample period from the time domain harmonic compression and expansion control means (102), and expanding the speech signal corresponding to one sample period to a plurality of sample periods.

Claim 8

A speech coding transmission equipment having a transmitting portion transmitting speech signal processed by a time domain harmonic compression and coded, and a receiving portion decoding the transmitted coded signal, and expanding and reproducing the decoded signal to a bandwidth existing before compression of the signal, said transmitting portion comprising, a time domain harmonic compression and expansion control means (102) distinguishing whether the speech signal is in the period of voiced speech or period of unvoiced speech, whereby, if the speech signal is the period of voiced speech a pitch period of the speech signal is extracted and output, and if the speech signal is in the period of unvoiced speech, a sample period signal by which the speech signal is sampled is output, and a time domain harmonic compression means (101) receiving the output of the pitch period from the time domain harmonic compression and expansion control means (102), obtaining the speech signal corresponding to one pitch period from the speech signal included in a plurality of pitch periods, and obtaining and outputting the speech signal corresponding to one sample period from the speech signal included in a plurality of sample periods receiving the output of the sample period from the time domain harmonic compression and expansion control means (102), and said receiving portion comprising, a time domain harmonic expansion means (106) receiving the output of the pitch period from the time domain harmonic compression and expansion control means (102), expanding the speech signal corresponding to one pitch period to a plurality of pitch periods, receiving the output of the sample period from the time domain harmonic compression and expansion control means (102), and expanding the speech signal corresponding to one sample period to a plurality of sample periods.