STATE DETECTING APPARATUS, COMMUNICATION APPARATUS, AND STORAGE MEDIUM STORING STATE DETECTING PROGRAM

• US 20130006630A1
• Filed: 04/13/2012
• Published: 01/03/2013
• Est. Priority Date: 06/30/2011
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: HAYAKAWA, Shoji, Matsuo, Naoshi
• Granted Patent: US 9,020,820 B2
• Time in Patent Office: Days
• US Class Current: 704/239
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
• CPC Class Codes: G10L 25/24 the extracted parameters being the cepstrum
• CPC Class Codes: G10L 25/63 for estimating an emotional state
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals

Abstract

A state detecting apparatus includes: a processor to execute acquiring utterance data related to uttered speech, computing a plurality of statistical quantities for feature parameters regarding features of the utterance data, creating, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities, computing a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data, and determining, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state; and a memory.

Claims

1. A state detecting apparatus comprising:

a processor to execute

acquiring utterance data related to uttered speech,

computing a plurality of statistical quantities for feature parameters regarding features of the utterance data,

creating, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities,

computing a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data, and

determining, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state; and

a memory to store the reference utterance data.

2. The state detecting apparatus according to claim 1,

wherein the processor computes the plurality of statistical quantities by taking the fundamental frequency and logarithmic power of the utterance data to be the feature parameters.

3. The state detecting apparatus according to claim 2,

wherein the processor individually computes a time average and a standard deviation for each of the feature parameters as the statistical quantities.

4. The state detecting apparatus according to claim 3,

wherein the processor creates pseudo-utterance data having a fundamental frequency whose time average or standard deviation is equal to the time average or standard deviation of the fundamental frequency in a plurality of frames included in the utterance data.

5. The state detecting apparatus according to claim 3,

wherein the processor creates pseudo-utterance data having a logarithmic power whose time average or standard deviation is equal to the time average or standard deviation of the logarithmic power in a plurality of frames included in the utterance data.

6. The state detecting apparatus according to claim 3,

wherein the processor determines whether the speaker is in the first state or the second state on the basis of the difference between the synthetic utterance data and the reference utterance data in a two-dimensional space defined by the standard deviation of the fundamental frequency and the standard deviation of the logarithmic power.

7. The state detecting apparatus according to claim 6, wherein

in the case where the difference is less than a first threshold value, the processor determines that the speaker is in the first state which resembles the state when the other uttered speech was uttered, and

in the case where the difference is equal to or greater than the first threshold value, the processor determines that the speaker is in the second state which does not resemble the state when the other uttered speech was uttered.

8. The state detecting apparatus according to claim 1,

wherein the processor determines whether the speaker is in the first state or the second state while removing statistical quantities larger than a second threshold value from among the statistical quantities for a plurality of frames included in the utterance data.

9. The state detecting apparatus according to claim 1,

wherein the processor additionally executes

updating the reference utterance data to other reference utterance data on the basis of the utterance data.

10. The state detecting apparatus according to claim 9,

wherein the processor updates to the other reference data using the reference data in the case where the difference between the statistical quantities of the utterance data and the statistical quantities of the reference utterance data is less than a third threshold value.

11. A communication apparatus that communicates with other communication apparatus, comprising:

a processor to execute

acquiring utterance data related to uttered speech,

computing a plurality of statistical quantities for feature parameters regarding features of the utterance data,

creating, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities,

computing a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data,

determining, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state, and

outputting the determination result; and

a memory to store the reference utterance data.

12. A state detecting apparatus comprising:

an acquirer configured to acquire utterance data related to uttered speech;

a first computation unit configured to compute a plurality of statistical quantities for feature parameters regarding features of the utterance data;

a pseudo-utterance data creator configured to create, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities;

a second computation unit configured to compute a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data; and

a detector configured to determine, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state.

13. A storage medium storing a state detecting program that causes a computer to execute operations, the operations comprising:

acquiring utterance data related to uttered speech,

computing a plurality of statistical quantities for feature parameters regarding features of the utterance data,

creating, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities,

computing a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data, and

determining, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state.