Speech packet communication system
First Claim
1. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
- means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal;
encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations;
.packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means;
means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or property of the input speech signal;
means connected to said prediction coefficient calculating means for generating the prediction signal; and
first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal;
wherein said first detection means comprises second detection means for detecting a prediction gain of the frame period of the differential speech signal and comparing the detected prediction gain with a predetermined threshold;
wherein the number of differential coded bits of said encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are determined depending on the result of the detection by said second detection means;
wherein when the detected prediction gain is equal to, or higher than, the predetermined threshold, the number of differential encoded bits of the encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are relatively large, and wherein when the detected prediction gain is smaller than the predetermined threshold, the number of differential encoded bits of the encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are relatively small.
1 Assignment
0 Petitions
Accused Products
Abstract
In a speech packet transmission system, a transmitter (speech terminal) first linearly predicts a speech signal input at a predetermined sample period from the input signal in the past and outputs the difference between the input speech signal and the prediction signal. A prediction coefficient used for linear prediction at this time is adapted with an output differential coded value at each predetermined sample period such that the prediction value is as close as possible to the value of the input waveform. The difference between the input speech signal and the prediction signal is converted to a differential code. A packet is prepared from a plurality of differential codes obtained during a frame period concerned and transmitted to a transmission line. A bit accuracy indicative of the differential code and a bit accuracy indicative of the differential coded value used for adaptation of the prediction coefficient are changed depending on the nature or property of the input speech signal.
27 Citations
8 Claims
-
1. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
-
means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal; encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations;
.packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means; means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or property of the input speech signal; means connected to said prediction coefficient calculating means for generating the prediction signal; and first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal; wherein said first detection means comprises second detection means for detecting a prediction gain of the frame period of the differential speech signal and comparing the detected prediction gain with a predetermined threshold; wherein the number of differential coded bits of said encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are determined depending on the result of the detection by said second detection means; wherein when the detected prediction gain is equal to, or higher than, the predetermined threshold, the number of differential encoded bits of the encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are relatively large, and wherein when the detected prediction gain is smaller than the predetermined threshold, the number of differential encoded bits of the encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are relatively small.
-
-
2. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
-
means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal; encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations; packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means; means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or property of the input speech signal; means connected to said prediction coefficient calculating means for generating the prediction signal; and first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal; wherein said packet assembling means comprises a first packet assembler for preparing a first packet of the respective most significant bits of a plurality of differential coded speech signals obtained during the frame period and a second packet assembler for preparing a second packet of the respective least significant bits of the plurality of differential encoded speech signals obtained during the frame period; wherein each terminal means comprises priority assignment means for determining first and second priority indicators indicative of the priorities of transmission of the first and second packets to be incorporated into the first and second packets in accordance with the result of detection of said first detection means; wherein said first detection means comprises second detection means for generating a parameter indicating whether the speech signal input during the frame period represents a speech or silence, and wherein when the parameter represents a speech, the first priority is assigned to the first packet and the second priority is assigned to the second packet while when the parameter represents silence, the first priority is assigned to the first packet and the second packet is refrained from transmission. - View Dependent Claims (3, 4, 5)
-
-
6. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
-
means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal; encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations; packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means; means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or property of the input speech signal; means connected to said prediction coefficient calculating means for generating the prediction signal; and first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal; wherein the differential encoded speech signal output during the frame period comprises at least six bits.
-
-
7. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
-
means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal; encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations; packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means; means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or properly of the input speech signal; means connected to said prediction coefficient calculating means for generating the prediction signal; and first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal; wherein the number of differential encoded bits used for adaptation of the prediction coefficient in the frame period comprises at least three bits.
-
-
8. A speech packet communication system in which two terminals communicate speech signal packets through transmission line means, each said terminal means comprising:
-
means for generating a differential speech signal indicative of the difference between an input speech signal inputted at a predetermined sample period and its prediction signal; encoding means for converting the differential speech signal to a differential coded speech signal of a plurality of bits at each predetermined sample period, the number of bits of the differential coded speech signal varying depending on the nature or property of the speech signal input during a frame period comprising a plurality of sample period durations; packet assembling means for preparing a packet from the plurality of differential coded speech signal obtained in the frame period and delivering the packet to said transmission line means; means for calculating a prediction coefficient from the differential coded speech signal, the number of differential coded bits used for adaptation of the prediction coefficient varying depending on the nature or property of the input speech signal; means connected to said prediction coefficient calculating means for generating the prediction signal; first detection means for detecting the nature or property of the input speech signal inputted in the frame period from one of the input speech signal and the differential speech signal; wherein said first detection means comprises second detection means for detecting a prediction gain of the frame period of the differential speech signal and comparing the detected prediction gain with a predetermined threshold; wherein the number of differential coded bits of said encoding means and the number of differential encoded bits used for adaptation of the prediction coefficient are determined depending on the result of the detection by said second detection means; and means for changing the predetermined threshold value for the prediction signal in accordance with the speech quality required by the user.
-
Specification