SPEECH PROCESSING DEVICE AND SPEECH PROCESSING METHOD

• US 20130144622A1
• Filed: 09/14/2011
• Published: 06/06/2013
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• Current Assignee: Panasonic Intellectual Property Management Co., Ltd. (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Intellectual Property Management Co., Ltd. (Panasonic Holdings Corporation)
• Inventors: Yamada, Maki, Endo, Mitsuru
• Granted Patent: US 9,064,501 B2
• Time in Patent Office: Days
• US Class Current: 704/248
• CPC Class Codes: G10L 2021/02087 the noise being separate speech, e.g. cocktail party
• CPC Class Codes: G10L 2021/065 Aids for the handicapped in understanding
• CPC Class Codes: G10L 2025/783 based on threshold decision
• CPC Class Codes: G10L 25/00 Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 muting semiconductor-based amplifiers when some special characteristics of a signal are sensed by a speech detector, e.g. sensing when no signal is present, H03G3/34
• CPC Class Codes: G10L 25/06 the extracted parameters being correlation coefficients
• CPC Class Codes: G10L 25/48 specially adapted for particular use
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10
• CPC Class Codes: H04R 2225/43 Signal processing in hearing aids to enhance the speech intelligibility
• CPC Class Codes: H04R 25/407 Circuits for combining signals of a plurality of transducers
• CPC Class Codes: H04R 25/552 Binaural
• CPC Class Codes: H04R 25/558 Remote control, e.g. of amplification, frequency

Abstract

A speech processing device which can accurately extract a conversation group from among a plurality of speakers, even when a conversation group formed of three or more people is present. This device (400) comprises: a spontaneous speech detection unit (420) and a direction-specific speech detection unit (430) which separately detect, from a sound signal, uttered speech from the speakers; a conversation establishment level calculation unit (450) which calculates a conversation establishment level for each separated segment of the time being determined, for all of the pairings of two people, on the basis of the detected uttered speech; an extended-period characteristic amount calculation unit (460) which calculates an extended-period characteristic amount for the conversation establishment level of the time being determined, for each pairing; and a conversation-partner determination unit (470) which extracts a conversation group which forms a conversation on the basis of the calculated extended-period characteristic amount.

Claims

1. A speech processing device, comprising:

a speech detector that detects speech of individual speakers from acoustic signals;

an established-conversation calculator that calculates degrees of established conversation of all pairs of the speakers in individual segments defined by dividing a determination time period, on the basis of the detected speech;

a long-time feature calculator that calculates a long-time feature of the degrees of established conversation within the determination time period for each of the pairs; and

a conversational-partner determining unit that extracts a conversation group holding conversation from the speakers, on the basis of the calculated long-time feature.

2. The speech processing device according to claim 1, wherein the degree of established conversation is a value indicating a rate of a time when one of the two speakers gives speech and the other gives no speech.

3. The speech processing device according to claim 1, further comprising

a total-amount-of-speech calculator that calculates total amounts of speech of all the respective pairs in each of the segments, the amount being sum of amounts of speech of the speakers, wherein

the established-conversation calculator invalidates the degree of established conversation in the segment having the total amount of speech smaller than a predetermined threshold, in calculation of the long-time feature.

4. The speech processing device according to claim 1, wherein

the acoustic signals are acoustic signals of speech received by a speech receiving section having variable directivity, the speech receiving section being disposed close to a user being one of the speakers, and

the device further comprises an output sound controller that controls the directivity of the speech receiving section toward one of the speakers other than the user of the conversation group if the extracted conversation group includes the user.

5. The speech processing device according to claim 4, wherein

the output sound controller performs predetermined signal processing on the acoustic signals and outputs the acoustic signals after the predetermined signal processing to a speaker of a hearing aid on the user.

6. The speech processing device according to claim 4, wherein

the speech detector detects speech of a speaker sitting in each of predetermined directions relative to the user, and

the output sound controller controls the directivity of the speech receiving section toward one of the speakers other than the user in the extracted conversation group.

7. The speech processing device according to claim 1, wherein

if the long-time features are uniformly high in several pairs of all the pairs, the conversational-partner determining unit determines that the speakers of the several pairs belong to the same conversation group.

8. The speech processing device according to claim 1, wherein

if a difference between the highest long-time feature and the second highest long-time feature is equal to or greater than a predetermined threshold in a pair including a user, the conversational-partner determining unit determines a speaker other than the user corresponding to the highest long-time feature to be an only conversational partner of the user.

9. The speech processing device according to claim 1, wherein the determination time period is a period from the last start of conversation in which the user participates to a current time.

10. A speech processing method, comprising:

detecting speech of individual speakers from acoustic signals;

calculating degrees of established conversation of all pairs of the speakers in individual segments defined by dividing a determination time period, on the basis of the detected speech;

calculating a long-time feature of the degrees of established conversation within the determination time period for each of the pairs; and

extracting a conversation group holding conversation from the speakers on the basis of the calculated long-time feature.