Keypad-assisted speech recognition for text or command input to concurrently-running computer application

• US 5,937,380 A
• Filed: 06/26/1998
• Issued: 08/10/1999
• Est. Priority Date: 06/27/1997
• Status: Expired due to Fees

• Current Assignee: Segan LLC
• Original Assignee: M.H. Segan Limited Partnership
• Inventors: Segan, Marc H.
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Smits, Talivaldis Ivars
• Application Number:
  US09/105,662
• Time in Patent Office: 410 Days
• Field of Search: 704/235, 704/252, 704/254, 704/275
• US Class Current: 704/235
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/24 Speech recognition using non-acoustical features
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A method for recognizing a spoken audio signal as input data to a computer includes receiving the spoken audio signal at an input device of the computer, receiving a typed first letter of the spoken audio signal at the input device of the computer and searching through entries that begin with the typed first letter in a dictionary in the computer for a best match with the spoken audio signal.

Claims

1. A method for facilitating the identification of spoken words by a speech recognition computer application which accesses a storage device containing dictionary entries and which inputs identified words into a concurrently running computer application, the method comprising the steps of:

a. receiving an audio signal representing a word spoken by a user;

b. receiving a user keyed entry through a user-manipulable key pad representing at least the initial letter of the spoken word;

c. after steps a. and b., automatically selecting a match between said spoken word and said dictionary entries by comparing said audio signal only with those dictionary entries containing said at least the initial letter; and

e. then inserting said match into the concurrently running computer application, without requiring further user input.

2. The method of claim 1, wherein said steps a. and b. are performed substantially simultaneously.

3. The method of claim 1, wherein one of said steps a. and b. is performed before the other of said steps a. and b., and said computer interface method further comprises the step of storing said one of said steps a. and b. in a buffer register until the other of said steps a. and b. is performed.

4. The method of claim 1, further comprising the steps of:

f. determining whether the spoken audio signal comprises a predetermined function;

g. performing the predetermined function if it is determined in step f. that the spoken audio signal comprises the predetermined function; and

h. performing step c. if the spoken audio interface is not the predetermined function.