Voice detecting and voice control system

• US 6,574,592 B1
• Filed: 03/20/2000
• Issued: 06/03/2003
• Est. Priority Date: 03/19/1999
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Saito, Masaki, Nankawa, Keigo
• Primary Examiner(s): Banks-Harold, Marsha D.
• Assistant Examiner(s): Lerner, Martin
• Application Number:
  US09/531,445
• Time in Patent Office: 1,170 Days
• Field of Search: 704/205, 704/206, 704/207, 704/208, 704/210, 704/214, 704/215, 704/246, 704/248, 704/270, 704/275, 708/300, 708/320, 708/309, 379/80
• US Class Current: 704/206
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2025/783 based on threshold decision
• 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

The voice detecting device includes a voice frequency detector for detecting the frequency of the input voice, then discriminating whether the detected frequency falls within a preset frequency range of voices to be detected, and outputting the result of the discrimination; an input signal level detector for detecting the energy level of the input voice, then comparing it to confirm whether the detected energy level exceeds a preset energy level threshold value of voices to be detected, and outputting the result of the comparison; a voice input judge part responsive to the result of discrimination and the result of comparison to judge whether a voice satisfying conditions for voice detection has been input or not, and output a first status signal in accordance with the result of the judgement; and a voice duration measure part for measuring the duration of the first status signal, then comparing it to confirm whether the detected duration falls within a preset range of duration threshold value, and outputting a second status signal in accordance with the result of the comparison. The voice control system includes a microphone for introducing a voice, the voice detecting device according to the invention, a control means responsive to one or both of the first status signal and second status signal to determine whether to output a control command, and a controlled objective which executes an operation according to the control command.

Claims

1. A voice control system comprising:

a microphone for introducing a voice;

a voice detecting device which includes a voice frequency detector for detecting the frequency of the input voice, then discriminating whether the detected frequency falls within a preset frequency range of voices to be detected, and outputting the result of the discrimination;

an input signal level detector for detecting the energy level of the input voice, then comparing it to confirm whether the detected energy level exceeds a preset energy level threshold value of voices to be detected, and outputting the result of the comparison;

a voice input judge part responsive to the result of discrimination by said voice frequency detector and the result of comparison by said input signal level detector to judge whether a voice satisfying conditions for voice detection has been input or not, and output a first status signal in accordance with the result of the judgement; and

a voice duration measure part for measuring the duration of said first status signal, then comparing it to confirm whether the detected duration falls within a preset range of duration threshold value, and outputting a second status signal in accordance with the result of the comparison;

a control means responsive to one or both of said first status signal and said second status signal to determine whether to output a control command; and

a controlled objective which executes an operation according to said control command,

wherein said voice frequency detector is a digital filter including a band pass filter block for setting said frequency range of voices to be detected and supplied with a voice, and an energy discriminator for identifying the voice from a signal coming through said band pass filter block,

wherein said band pass filter block is a cascade-connected low pass filter block and high pass filter block, and

wherein said low pass filter block is a cascade-connected quadratic IIR filters of three stages, and said high pass filter block is a quadratic IIR filter of one stage.