Electronic device and method

Patent metadata

Field	Value
Publication number	US-10770077-B2
Application number	US-201916298889-A
Country	US
Kind code	B2
Filing date	Mar 11, 2019
Priority date	Sep 14, 2015
Publication date	Sep 8, 2020
Grant date	Sep 8, 2020

Abstract

According to one embodiment, an electronic device records an audio signal, determines a plurality of user-specific utterance features within the audio signal, the plurality of user-specific utterance features including a first set of user specific-utterance features associated with the registered user and a second set of user-specific utterance features associated with the unregistered user, and displays the identifier of the registered user differently than an identifier of the unregistered user.

First claim

What is claimed is:

1. An electronic device comprising:
a microphone configured to collect a voice;
a memory in which at least the collected voice and speaker feature data are stored;
a display configured to display at least a recording view and a reproduction view as a display screen; and
a hardware processor configured to execute a voice recorder application, the hardware processor configured to:
record the voice collected by the microphone as audio data on the memory according to a recording operation in the recording view;
classify a plurality of voice sections of the recorded audio data into a plurality of clusters corresponding to a plurality of speakers, for displaying the reproduction view;
extract a speaker feature amount included in one or more voice sections classified into the same cluster, and register in the memory the speaker feature amount as the speaker feature data;
delete the speaker feature data whose importance is low if the number of the speaker feature data of the memory exceeds a predetermined number; and
compare the extracted speaker feature amount with the speaker feature data which have been registered, and identify a speaker name included in the speaker feature data which includes the extracted speaker feature amount as a speaker of the voice sections.

2. The electronic device of claim 1 , wherein the hardware processor is further configured to:
acquire the number of times the speaker feature amount has been identified up to the present as information on importance, if no speaker name is included in the speaker feature data;
identify the speaker name as a speaker who appeared in the past but whose speaker name has not been registered yet, if the number of times is greater than or equal to two; and
identify the speaker name as a new speaker who did not appear in the past, if the number of times is one.

3. The electronic device of claim 1 , wherein speaker feature provisional data including a provisionally-registered speaker feature amount and a speaker name is further stored in the memory.

4. The electronic device of claim 3 , wherein the hardware processor is further configured to:
acquire, if a speaker name corresponding to one or more voice sections classified into a predetermined cluster is added in the reproduction view, the speaker feature amount corresponding to the voice sections; and
generate speaker feature provisional data including the acquired speaker feature amount and a speaker name input from the display screen.

5. The electronic device of claim 4 , wherein the hardware processor is further configured to register, when the extracted speaker feature amount is registered in the memory as the speaker feature data and if the speaker feature provisional data is stored in the memory, the speaker feature provisional data.

6. The electronic device of claim 1 , wherein importance related to a speaker feature amount is added to the speaker feature data, and
the hardware processor is further configured to delete, when the extracted speaker feature amount is registered in the memory as the speaker feature data and if the number of speaker feature data registered in the memory is greater than or equal to the predetermined number, the speaker feature data whose importance is low until the number of speaker feature data becomes less than the predetermined number.

7. The electronic device of claim 6 , wherein the importance is updated each time the speaker feature amount in the speaker feature data is identified as being included in the recorded audio data.

8. The electronic device of claim 1 , wherein the hardware processor is further configured to display all user names registered in the memory with the speaker feature amounts in a selectable form as a correction candidate in response to a command to correct a speaker name from the display screen.

9. The electronic device of claim 1 , wherein the hardware processor is further configured to display on the display screen a tutorial for inputting a speaker name, if all statuses of speaker names displayed in the reproduction view are new speakers.

10. A method of a display device comprising:
a microphone configured to collect a voice;
a memory in which at least the collected voice and speaker feature data are stored; and
a display configured to display at least a recording view and a reproduction view as a display screen, the method comprising:
recording the voice collected by the microphone as audio data on the memory according to a recording operation in the recording view;
classifying a plurality of voice sections of the recorded audio data into a plurality of clusters corresponding to a plurality of speakers, for displaying the reproduction view;
extracting a speaker feature amount included in one or more voice sections classified into the same cluster, and register in the memory the speaker feature amount as the speaker feature data;
deleting the speaker feature data whose importance is low if the number of the speaker feature data of the memory exceeds a predetermined number; and
comparing the extracted speaker feature amount with the speaker feature data which have been registered, and identifying a speaker name included in the speaker feature data which includes the extracted speaker feature amount as a speaker of the voice sections.

Assignees

• Toshiba Client Solutions Co Ltd

Inventors

• Kikugawa Yusaku

Classifications

• G06F3/165 Primary

  Management of the audio stream, e.g. setting of volume, audio stream path · CPC title

• G10L17/00

  Speaker identification or verification techniques · CPC title

• G10L25/87

  Detection of discrete points within a voice signal · CPC title

• G10L21/12

  by displaying time domain information · CPC title

• G10L17/22 Primary

  Interactive procedures; Man-machine interfaces · CPC title