System and method for speech recognition and transcription

• US 6,990,445 B2
• Filed: 12/17/2001
• Issued: 01/24/2006
• Est. Priority Date: 12/17/2001
• Status: Expired due to Term

• Current Assignee: LYDA HUNT - BUNKER TRUST - MARY M. HUNT
• Original Assignee: XL8 Systems Incorporated
• Inventors: Ky, Dung H.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): HARPER, V PAUL
• Application Number:
  US10/022,947
• Publication Number:
  US 20030130843A1
• Time in Patent Office: 1,499 Days
• Field of Search: 704/275, 704/256, 704/251, 704/239, 704/235, 704/231, 704/200, 704/1
• US Class Current: 704/235
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 2015/027 Syllables being the recognition units
• CPC Class Codes: G10L 2015/088 Word spotting

Abstract

The present invention comprises a method and system for speech recognition. The method includes the steps of receiving digital data representation of speech having at least one word, which includes a waveform representation of the at least one word. The method includes searching a library containing digital data representation of a plurality of words and matching the digital data representation of the at least one word to digital data representation of a word in the library. The method further includes the steps of determining the number of syllables in the digital data representation of the at least one word, and then providing an ASCII representation of the matched word in response to the number of syllables of the at least one word being identical to the number of syllables of the matched word in the library.

Claims

1. A method for speech recognition, comprising:

receiving a digital data representation of speech having at least one word as a waveform and a digital binary bit stream;

analyzing the digital data waveform representation for at least one parameter and comparing the at least one parameter with a parameter of digital data representations corresponding to words stored in a library;

determining a set of ASCII characters based on the analyzed digital data waveform representation;

determining the number of syllables in the speech having the at least one word based on the digital data waveform representation;

determining whether the set of ASCII characters corresponds to the speech having the at least one word by matching the number of syllables determined for the digital data waveform representation with the number of syllables for the word stored in the library;

if the number of syllables do not match,

grouping the digital representation of speech binary bit stream into subsets;

mapping each subset of the digital representation of speech into a character representation of speech and generating a stream of character representations from the digital representation;

parsing the stream of character representations to group the character representations of speech into words.

2. The method as set forth in claim 1, further comprising:

receiving specific input having at least some of the ASCII characters corresponding to the mapping of the digital data representation of the speech having the at least one word;

storing the input; and

associating the input including at least some ASCII characters with the received digital data representation of the speech having the at least one word.

3. The method as set forth in claim 2 wherein the input is received from a keyboard.

4. The method as set forth in claim 2 wherein the input received is user auditory input from a sound card.

5. The method as set forth in claim 1, further comprising matching the digital data representation of the speech having the at least one word as a waveform to a digital data representation in the library based on at least one of:

waveform frequency, period and amplitude.

6. A speech recognition system, comprising:

a digital data representation of speech having at least one word;

an interpreter operable to analyze the waveform a digital data representation of speech having at least one word, determine a set of ASCII characters based on the analyzed digital data representation;

a syllable matching process operable to analyze the digital data representation waveform of the speech having the at least one word, determine the number of syllables in each word in the of speech having the at least one word, and determine whether the set of ASCII characters corresponds to the speech having the at least one word by matching the determined number of syllables with the number of syllables for the word selected from the library;

if the number of syllables do not correspond,

grouping the digital representation of speech binary bit stream into subsets;

mapping each subset of the digital representation of speech into a character representation of speech and generating a stream of character representations from the digital representation; and

parsing the stream of character representations to group the character representations of speech into words; and

a display screen operable to display the set of ASCII characters.