Voice information control method and terminal device

Patent metadata

Field	Value
Publication number	US-2016019894-A1
Application number	US-201514794847-A
Country	US
Kind code	A1
Filing date	Jul 9, 2015
Priority date	Jul 16, 2014
Publication date	Jan 21, 2016
Grant date	—

Abstract

A voice information control method for a terminal used in a system including a server device which creates text data on the basis of the voice information received from the terminal device, the method including: acquiring plurality items of first voice information; specifying a time interval that includes second voice information which is one of the plurality items of the first voice information, and which includes spoken voice of a first speaker who uses the first terminal device; and transmitting the second voice information included in the specified time interval being transmitted to the server device.

First claim

What is claimed is:

1 . A voice information control method for a terminal device used in a system including a server device which creates text data in accordance with the voice information received from the terminal device, the voice information control method comprising:
acquiring plurality items of first voice information; specifying a time interval that includes second voice information which is one of the plurality items of the first voice information, and which includes spoken voice of a first speaker who uses the terminal device; and transmitting the second voice information included in the specified time interval to the server device.

2 . The voice information control method according to claim 1 ,
wherein the terminal device retains a voice model including voice information relating to the first speaker, and the specifying is performed by comparing the plurality items of the first voice information and the voice model.

3 . The voice information control method according to claim 1 ,
wherein only the second voice information included in the specified time interval is transmitted to the server device in the transmitting, and the plurality items of the first voice information included in a time interval other than the specified time interval is not transmitted to the server device in the transmitting.

4 . The voice information control method according to claim 1 , further comprising:
accumulating the second voice information included in the specified time interval into a recording medium in the first terminal device, wherein the accumulated second voice information is transmitted to the server device at a prescribed timing in the transmitting.

5 . The voice information control method according to claim 4 ,
wherein the prescribed timing is a timing at which the plurality items of the first voice information acquired by the acquiring have become equal to or less than a prescribed sound pressure level.

6 . The voice information control method according to claim 4 ,
wherein the prescribed timing is a timing in a time interval other than the specified time interval that includes the second voice information.

7 . The voice information control method according to claim 4 ,
wherein the prescribed timing is a timing at which the first speaker has instructed the system to terminate by way of the terminal device which the first speaker uses.

8 . The voice information control method according to claim 1 ,
wherein at least one of the acquiring, the specifying, and the transmitting is carried out by a processor in the terminal device.

9 . A voice information control method for a server device in a system which creates text data in accordance with voice information received from each of a plurality of terminal devices, the voice information control method comprising:
receiving plurality items of the voice information and time information regarding times at which the plurality of items of voice information are acquired by each of the plurality of terminal devices; carrying out voice recognition and text conversion with respect to each of the plurality of items of voice information to thereby create plurality items of text data; and integrating the plurality of items of text data in a time-sequential manner in accordance with the time information.

10 . The voice information control method according to claim 9 , further comprising:
transmitting the integrated plurality of items of text data to some of the terminal devices or all of the terminal devices from among the plurality of terminal devices.

11 . The voice information control method according to claim 10 ,
wherein plurality items of first voice information, corresponding to the plurality of items of the voice information received from a first terminal device from among the plurality of terminal devices, include second voice information which is one of the plurality items of the first voice information and which includes spoken voice of a first speaker who uses the first terminal device.

12 . The voice information control method according to claim 10 ,
wherein the plurality of items of the voice information received from each of the plurality of terminal devices constitute all of the voice information which each of the plurality of terminal devices acquired while each of the terminal devices is activating the system.

13 . The voice information control method according to claim 12 , further comprising:
specifying the terminal device used by a speaker who spoke in each speech interval by comparing each of the plurality of items of voice information received from each of the plurality of terminal devices.

14 . The voice information control method according to claim 13 , the specifying further comprising:
comparing sound pressure levels of each of the plurality of items of voice information received from each of the plurality of terminal devices; and specifying the terminal device that has a highest sound pressure level in a first speech interval as the terminal device used by the speaker in the first speech interval.

15 . The voice information control method according to claim 12 , further comprising:
specifying second voice information which is one of plurality items of first voice information, as the plurality items of the voice information received from a first terminal device from among the plurality of terminal devices, and which includes spoken voice of a first speaker who uses the first terminal device by comparing the plurality items of the first voice information and the plurality of items of the voice information received from a second terminal device from among the plurality of terminal devices; specifying one of the plurality items of the voice information, received from a second terminal device, which does not include spoken voice of a second speaker who uses the second terminal device as noise voice; and removing the specified noise voice from the first voice information before the voice recognition and text conversion are carried out using the second voice information.

16 . The voice information control method according to claim 10 , further comprising:
creating minutes associated with the terminal devices which have acquired the plurality items of the voice information, the time information, and the text-converted text data; and transmitting the created minutes to some of the terminal devices or all of the terminal devices from among the plurality of terminal devices which the voice information have acquired.

17 . The voice information control method according to claim 9 ,
wherein at least any one of the receiving, the creating, and the integrating is carried out using a processor.

18 . A terminal device used in a system including a server device which creates text data in accordance with the voice information received from the plurality of terminal devices, the terminal device comprising:
a voice acquisitioner that acquires first voice information; a controller/processor that specifies a time interval that includes second voice information which is one of the plurality items of the first voice information, and which includes spoken voice of a first speaker who uses the terminal device; and a communicator that transmits the second voice information included in the specified time interval to the server device.

19 . The terminal device according to claim 18 ,
wherein the terminal device comprises a recorder that retains a voice model that includes voice information relating to the first speaker, an

Assignees

• Panasonic Ip Corp America

Inventors

• Tsujikawa Misaki
• Shimada Isao

Classifications

• G10L17/00

  Speaker identification or verification techniques · CPC title

• G10L15/26 Primary

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G10L15/30

  Distributed recognition, e.g. in client-server systems, for mobile phones or network applications · CPC title

• H04M2250/74

  with voice recognition means · CPC title

• G10L2021/02087

  the noise being separate speech, e.g. cocktail party · CPC title