Speaking period detection device, voice recognition processing device, transmission system, signal level control device and speaking period detection method

• US 7,627,470 B2
• Filed: 09/14/2004
• Issued: 12/01/2009
• Est. Priority Date: 09/19/2003
• Status: Expired due to Fees

• Current Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Original Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Inventors: Sugimura, Toshiaki, Miki, Toshio, Hayashi, Kouki, Hiraiwa, Akira, Manabe, Hiroyuki, Hiraiwa, legal representative
• Primary Examiner(s): Hudspeth; David R
• Assistant Examiner(s): Rider; Justin W
• Application Number:
  US10/939,566
• Publication Number:
  US 20050102134A1
• Time in Patent Office: 1,904 Days
• Field of Search: 600/23, 600/382
• US Class Current: 704/231
• CPC Class Codes: G06F 3/015 Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
• CPC Class Codes: G10L 15/25 using position of the lips, movement of the lips or face analysis
• 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

Abstract

A speaking period required for a voice recognition processing is detected with a simple and robust approach. A speaking period is detected based on an EMG generated when a speaker speaks. A large amplitude is observed in an EMG as muscular activity is caused when a speaker speaks. By observing this, a speaking period can be detected. The EMG can be measured from the speaker'"'"'s skin via electrodes provided on a mobile-type terminal. Since a mobile-type terminal is usually pressed to the skin for use, affinity between this use form and the present invention is very high.

Claims

1. A speaking period detection device comprising:

an electromyography (EMG) detection unit for detecting an EMG from a pair of electrodes provided on a portion of the case of a terminal used by a speaker for speaking, the portion being in contact with the speaker'"'"'s skin while the speaker is speaking;

a speaking period detection unit for detecting a speaking period a value obtained from a feature amount extracted from the EMG detected by the EMG detection unit with a predetermined threshold to detect a speech start timing and a speech end timing of the speaking period; and

a counter for counting the result of the comparison of the value obtained from the feature amount with the predetermined threshold,

wherein the speaking period detection unit determines the speaking period only when a count value of the counter exceeds a predetermined value, and

wherein the feature amount is any one of a root mean square (RMS) value, an average relative variation (ARV) value, an integrated electro muscle graphy (iEMG), a frequency spectrum of an EMG, and the number of crossings with a threshold value.

2. The speaking period detection device according to claim 1, further comprising a storage unit for saving a speaking voice signal of the speaker;

wherein saving of the speaking voice signal is started at the speech start timing and ended at the speech end timing.

3. A voice recognition processing device comprising a voice recognition processing unit for performing a voice recognition processing for a speaking voice signal corresponding to the speaking period detected by the speaking period detection device according to claim 1.

4. A voice recognition processing device comprising a voice recognition processing unit for performing a voice recognition processing for a speaking voice signal stored in a storage unit included in the speaking period detection device according to claim 1.

5. A transmission system for performing a coding processing for a speaking voice signal corresponding to the speaking period and transmitting the coded voice signal, the transmission system comprising a coding processing timing determination unit for determining timing of the coding processing based on the speaking period detected by the speaking period detection unit according to claim 1.

6. A signal level control device for controlling a level of an output signal derived from a microphone, the signal level control device comprising a control unit for controlling the level of the output signal based on the speaking period detected by the speaking period detection unit according to claim 1, and controlling a voice corresponding to a non-speaking period not to be outputted.

7. A speaking period detection method comprising:

detecting an electromyography (EMG) from a pair of electrodes provided on a portion of the case of a terminal used by a speaker for speaking, the portion being in contact with the speaker'"'"'s skin while the speaker is speaking;

detecting a speaking period, by comparing a value obtained from a feature amount extracted from the EMG detected with a predetermined threshold to detect a speech start timing and a speech end timing of the speaking period; and

counting the result of the comparison of the value obtained from the feature amount with the predetermined threshold,

wherein the speaking period is determined only when a count value of the counter exceeds a predetermined value, and

wherein the feature amount is any one of a root mean square (RMS) value, an average relative variation (ARV) value, an integrated electro muscle graphy (iEMG), a frequency spectrum of an EMG, and the number of crossings with a threshold value.

8. A mobile terminal comprising:

a speaking period detection device; and

a pair of electrodes provided on a portion, of the mobile terminal, in contact with a facial skin surface of a speaker who is speaking,

the speaking period detection device comprising;

an electromyography (EMG) detection unit for detecting an EMG from the pair of electrodes;

a speaking period detection unit for detecting a speaking period, by comparing a value obtained from a feature amount extracted from the EMG detected by the EMG detection unit with a predetermined threshold to detect a speech start timing and a speech end timing of the speaking period; and

a counter for counting the result of the comparison of the value obtained from the feature amount with the predetermined threshold,

wherein the speaking period detection unit determines the speaking period only when a count value of the counter exceeds a predetermined value.

9. The mobile terminal according to claim 8, further comprising a voice activity detector (VAD) for determining presence or absence of a voice while the speaker is speaking,

wherein the detection of the speaking period based on an EMG signal and the detection of the speaking period based on a voice signal with the use of the VAD are integrated.