Voice input system

• US 4,641,342 A
• Filed: 03/19/1984
• Issued: 02/03/1987
• Est. Priority Date: 03/17/1983
• Status: Expired due to Term

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Watanabe, Takao, Watari, Masao
• Primary Examiner(s): Kemeny, E. S. Matt
• Application Number:
  US06/590,660
• Time in Patent Office: 1,051 Days
• Field of Search: 381/41-43, 381/46, 381/49, 364/466, 360/72.1
• US Class Current: 704/253
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 25/87 Detection of discrete points within a voice signal

Abstract

An input system for a voice recognizer circuit wherein a cue signal is issued to the user to indicate system readiness. A voice detector detects the presence of a voice signal. Control circuitry detects if a voice signal is detected prior to the end of the initial cue signal and, if so, causes a second cue signal to be issued, thereby preventing a loss of voice input. The voice detector can also be used to selectively switch an active one of a plurality of user channels to one of a smaller number of voice recognizers.

Claims

1. A voice input system, comprising:

an input cue signal generating means for producing a first cue signal informing the user that the system is ready to receive the voice input;

a voice detection means for producing a detection signal when the voice signal is being inputted; and

an input instructing means for ordering said input cue signal generating means to issue a second input cue signal when said detection signal exists prior to the end of said input cueing signal.

2. A voice input system as recited in claim 1, further comprising:

voice processing means for effecting predetermined processing on the inputted voice signal.

3. A voice signal input system according to claim 2, wherein said voice detection means includes means for determining a starting time t_s, when the level of the voice input signal has exceeded a predetermined level Th₁ at a moment t_s and then exceeded a predetermined level Th₂ thereafter, and for determining an ending time t_e, when the level of the voice input signal has come down below the level Th₁ at a moment t_e and stays below this level for a predetermined period thereafter.

4. A voice signal input system according to claim 2, wherein the voice processing means comprises means for recognizing the inputted voice signal.

5. A voice input system, comprising;

m (being an integer) separate user channels;

response means for delivering voice output to any of said user channels;

voice detection means for detecting the voice inputs on any of said user channels;

a change-over switch for selectively supplying m voice signals to any of n (n being an integer smaller than m) separate recognition input terminals;

processing means for processing the voices received from said n input terminals;

control means for connecting, only in the presence of at least one idle input terminal in the recognition means, said user channel in use to one of the idle input terminals and for issuing a voice input request; and

input instructing means for delivering, when the detecting means detects no voice in advance of the time at which the voice input request is received, a voice output to said user channel to inform the user that the system is ready to receive the voice input and for giving an instruction to start the operation of said processing means, the input instructing means further delivering, when the voice is detected by said detection means, a voice output calling upon the user to activate the voice input at the moment at which the ending point of the voice is detected, thereby starting the operation of said processing means.

6. A voice signal input system according to claim 5, wherein said voice detection means includes means for extracting voice characteristic information and power information of the inputted voice and for detecting the input of the voice using the power information, the characteristic information being delivered as input to said change-over switch.

7. A voice input system according to claim 6, wherein said voice detection means detects, when the level of the input voice signal has exceeded a predetermined consonant level Th₁ at a moment t_s and thereafter has exceeded a predetermined vowel level Th₂, the moment t_s as the beginning point and, when said input voice signal has fallen below Th₁ at a moment t_e and has stayed below Th₁ for a predetermined period thereafter, the moment t_e as the ending time.

8. A voice input system according to claim 5, wherein said voice processing means includes means for recognizing the input voice.