Claims
- 1. A speech recognition LSI system comprising:
- speech segment detecting means for receiving a speech signal and for detecting a speech segment corresponding to one word from the speech signal on the basis of an input level thereof, the speech segment being a section of the speech signal which has a level higher than a predetermined value over a predetermined period of time, and for generating start and end signals which indicate start and end portions of the detected speech segment;
- reference pattern memory means for storing reference patterns corresponding to words;
- speech recognition means for comparing a portion of the received speech signal which is defined as the one word by said speech segment detecting means, with the reference patterns stored in said reference pattern memory means and for selecting the reference pattern most similar to the speech segment detected by said speech segment detecting means; and
- recording/reproduction means for recording the speech signal and for reproducing, in accordance with the start and end signals generated from said speech segment detecting means, only the portion of the received speech signal which is defined as the one word by said speech segment detecting means.
- 2. The system according to claim 1, wherein said speech signal is an analog signal.
- 3. The system according to claim 2, further comprising speech analyzing means, provided preceding said speech segment detecting means, for converting the analog signal to a digital speech signal in accordance with the frequency of the analog signal and for delivering the digital speech signal as an output signal.
- 4. The system according to claim 3, wherein said speech segment detecting means detects a speech segment from the digital speech signal.
- 5. The system according to claim 2, wherein said speech segment and detects a speech segment from the analog speech signal.
- 6. The system according to claim 5, further comprising a speech analyzing means, provided subsequent to said speech segment detecting means, for converting an analog speech signal to a digital speech signal in accordance with the frequency of the analog speech signal and for delivering the digital speech signal as an output signal.
- 7. The system according to claim 2, wherein said recording/reproduction means records said analog speech signal.
- 8. The system according to claim 3, wherein said recording/reproduction means receives and records a digital speech signal supplied from said speech analyzing means.
- 9. The system according to claim 6, wherein said recording/reproduction means receives and records the digital speech signal supplied from said speech analyzing means.
- 10. The system according to claim 1, further comprising switch means for switching the output of said speech segment detecting means to one of said reference pattern memory means and said speech recognition means.
- 11. The system according to claim 6, further comprising switch means for switching the output of said speech segment detecting means to one of said reference pattern memory means and said speech recognition means.
Priority Claims (1)
Number |
Date |
Country |
Kind |
1-93604 |
Apr 1989 |
JPX |
|
Parent Case Info
This application is a continuation of application Ser. No. 07/507,773 filed Apr. 12, 1990, now abandoned.
US Referenced Citations (4)
Foreign Referenced Citations (3)
Number |
Date |
Country |
0077194A1 |
Apr 1983 |
EPX |
0178509A1 |
Apr 1986 |
EPX |
974850 |
Nov 1964 |
GBX |
Continuations (1)
|
Number |
Date |
Country |
Parent |
507773 |
Apr 1990 |
|