Speech recognition system including interrupt scheme that avoids operational and interfacing conflicts

• US 5,231,691 A
• Filed: 10/05/1990
• Issued: 07/27/1993
• Est. Priority Date: 10/06/1989
• Status: Expired due to Term

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Yasuda, Seigou
• Primary Examiner(s): Fleming, Michael R.
• Assistant Examiner(s): Doerrler, Michelle
• Application Number:
  US07/593,819
• Time in Patent Office: 1,026 Days
• Field of Search: 381/41-45, 395/2
• US Class Current: 704/200
• CPC Class Codes: G06F 3/16 Sound input; Sound output speech processing G10L
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

A speech recognition system in which speech information inputted from a microphone is recognized and a speech signal corresponding to the speech information is generated and converted into a set of characters equivalent to keystroke information normally inputted from a keyboard, the set of characters being supplied to a keyboard buffer, so that an application program running on a computer system executes a desired job. The speech recognition system comprises a first interrupt processing part which, after receiving an interrupt signal due to a depression of a key on the keyboard, checks that information within the keyboard buffer accords with a first predefined keystroke code, turns a first flag ON only when the information accords with the first predefined keystroke code and terminates a first interrupt procedure, and a second interrupt processing part which, after the first interrupt routine is terminated, enables a second interrupt into operation of the application program only when the first flag is turned ON, so that a speech information input procedure is started to provide the keystroke information thus recognized.

Claims

1. A speech recognition system comprising:

a) a computer system having;

1) a central processing unit for performing a prescribed operation;

2) a keyboard for inputting keystroke information to a keyboard buffer; and

3) a display for outputting information resulting from the prescribed operation by the central processing unit in a visible form;

the computer system being capable of executing an application program therein;

b) a microphone for inputting speech information to the speech recognition system and for supplying a speech signal corresponding to the speech information;

c) speech signal processing means for amplifying the speech signal and for correcting the amplified speech signal into a first signal;

d) feature parameter extraction means for analyzing a frequency of the first signal through a predetermined frequency analysis and for extracting a set of feature parameters from the first signal;

e) speech section detection means for detecting speech sections of the speech signal from the set of feature parameters extracted by said feature parameter extraction means by detecting a leading edge of said speech signal from said speech sections of said speech signal;

f) feature parameter input control means for supplying at prescribed time intervals the set of feature parameters extracted from the first signal to a speech data buffer;

g) a standardized pattern memory for storing standardized patterns;

h) speech recognition means for recognizing the speech information by comparing the set of feature parameters within said speech data buffer with the standardized patterns stored in the standardized pattern memory, said speech recognition means supplying a second signal corresponding to the recognized speech information;

i) outputting means for converting the second signal supplied from the speech recognition means into a set of characters and transferring the set of characters to the keyboard buffer;

j) first interrupt processing means for performing a first interrupt procedure after receiving a keyboard key depression interrupt signal due to a depression of a key on the keyboard, for detecting whether or not information within the keyboard buffer accords with a first predefined keystroke code, for turning a first flag ON only when the information accords with the first predefined keystroke code, and for terminating the first interrupt procedure; and

k) second interrupt processing means for performing a second interrupt procedure, the second interrupt processing means, only after the first interrupt procedure is terminated, for detecting whether or not the first flag is turned ON, and for enabling a second interrupt into the operation of the application program only when the first flag is turned ON, so that a speech information input procedure is started to provide the set of characters in replying to the speech information;

wherein;

A) immediately when the speech information input procedure is started, said feature parameter input control means is enabled to supply at said prescribed time intervals said set of feature parameters to said speech data buffer, and said interrupt signal is enabled to occur due to the depression of a key on the keyboard; and

B) after said speech section detection means detects a leading edge of the speech signal, an occurrence of a keyboard interrupt due to the depression of a key on the keyboard is disabled.

2. A speech recognition system as claimed in claim 1, wherein the first interrupt processing means and the second interrupt processing means are provided within the speech recognition system separately from the application program running on the computer system.

3. A speech recognition system as claimed in claim 1, wherein the first interrupt processing means and the second interrupt processing means are provided within the application program, the second interrupt routine being performed only when the application program is in a waiting state.

4. A speech recognition system as claimed in claim 3, wherein the second interrupt processing means checks that the first flag is turned ON only when the application program is in a waiting state.

5. A speech recognition system as claimed in claim 4, wherein the feature parameter input control means comprises a periodic signal generation part for supplying a periodic signal at said prescribed time intervals to the feature parameter input control means and to the speech data buffer, said prescribed time intervals at which the periodic signal generation part supplies the periodic signal being in a range between 5 ms and 20 ms.

6. A speech recognition system as claimed in claim 1, wherein;

a modification of the standardized patterns stored in the standardized pattern memory is allowed when a second predefined keystroke data is entered from the keyboard.

7. A speech recognition system, comprising:

a) a computer system having;

1) a central processing unit for performing a prescribed operation;

2) a keyboard for inputting keystroke information to a keyboard buffer; and

3) a display for outputting information resulting from the prescribed operation by the central processing unit in a visible form;

the computer system being capable of executing an application program therein;

b) a microphone for inputting speech information to the speech recognition system and for supplying a speech signal corresponding to the speech information;

c) speech signal processing means for amplifying the speech signal and for correcting the amplified speech signal into a first signal;

d) feature parameter extraction means for analyzing a frequency of the first signal through a predetermined frequency analysis and for extracting a set of feature parameters from the first signal;

e) speech section detection means for detecting speech sections of the speech signal from the set of feature parameters extracted by said feature parameter extraction means by detecting a leading edge of said speech signal from said speech sections of said speech signal;

f) feature parameter input control means for supplying at prescribed time intervals the set of feature parameters extracted from the first signal to a speech data buffer;

g) a standardized pattern memory for storing standardized patterns;

h) speech recognition means for recognizing the speech information by comparing the set of feature parameters within said speech data buffer with the standardized patterns stored in the standardized pattern memory, said speech recognition means supplying a second signal corresponding to the recognized speech information;

i) outputting means for converting the second signal supplied from the speech recognition means into a set of characters and transferring the set of characters to the keyboard buffer;

j) first interrupt processing means for performing a first interrupt procedure after receiving a keyboard key depression interrupt signal due to a depression of a key on the keyboard, for detecting whether or not information within the keyboard buffer accords with a first predefined keystroke code, turning a first flag ON only when the information accords with the first predefined keystroke code, and for terminating the first interrupt procedure; and

k) second interrupt processing means for performing a second interrupt procedure, the second interrupt processing means, only after the first interrupt procedure is terminated, for detecting whether or not the first flag is turned ON, and for enabling a second interrupt into the operation of the application program only when the first flag is turned ON, so that a speech information input procedure is started to provide the set of characters in replying to the speech information;

wherein a supplying of the set of feature parameters by the feature parameter input control means, and an occurrence of the interrupt signal due to the depression of a key on the keyboard, are each enabled immediately when the speech information input procedure is started; and

wherein a keyboard interrupt caused by the depression of a key on the keyboard is disabled after the speech section detection means detects a leading edge of the speech signal.

8. A speech recognition system, comprising:

a) a computer system having;

1) a central processing unit for performing a prescribed operation;

2) a keyboard for inputting keystroke information to a keyboard buffer; and

3) a display for outputting information resulting from the prescribed operation by the central processing unit in a visible form;

the computer system being capable of executing an application program therein;

b) a microphone for inputting speech information to the speech recognition system and for supplying a speech signal corresponding to the speech information;

c) speech signal processing means for amplifying the speech signal and for correcting the amplified speech signal into a first signal;

d) feature parameter extraction means for analyzing a frequency of the first signal through a predetermined frequency analysis and for extracting a set of feature parameters from the first signal;

e) speech section detection means for detecting speech sections of the speech signal from the set of feature parameters extracted by said feature parameter extraction means by detecting a leading edge of said speech signal from said speech sections of said speech signal;

f) feature parameter input control means for supplying at prescribed time intervals the set of feature parameters extracted from the first signal to a speech data buffer;

g) a standardized pattern memory for storing standardized patterns;

h) speech recognition means for recognizing the speech information by comparing the set of feature parameters within said speech data buffer with the standardized patterns stored in the standardized pattern memory, said speech recognition means supplying a second signal corresponding to the recognized speech information;

i) outputting means for converting the second signal supplied from the speech recognition means into a set of characters and transferring the set of characters to the keyboard buffer;

j) first interrupt processing means for performing a first interrupt procedure after receiving a keyboard key depression interrupt signal due to a depression of a key on the keyboard, for detecting whether or not information within the keyboard buffer accords with a first predefined keystroke code, for turning a first flag ON only when the information accords with the first predefined keystroke code, and for terminating the first interrupt procedure; and

k) second interrupt processing means for performing a second interrupt procedure, the second interrupt processing means, only after the first interrupt procedure is terminated, for detecting whether or not the first flag is turned ON, and for enabling a second interrupt into the operation of the application program only when the first flag is turned ON, so that a speech information input procedure is started to provide the set of characters in replying to the speech information;

wherein the feature parameter input control means includes a periodic signal generation part for supplying a periodic signal at said prescribed time intervals to the feature parameter input control means and to the speech data buffer, said prescribed time intervals at which the periodic signal generation part supplies the periodic signal being in a range of between 5 and 20 ms.