Syllable recognition system

• US 5,054,084 A
• Filed: 05/15/1990
• Issued: 10/01/1991
• Est. Priority Date: 04/05/1986
• Status: Expired due to Term

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Tanaka, Atsuo, Kamiya, Shin
• Primary Examiner(s): NOT, DEFINED
• Assistant Examiner(s): NOT, DEFINED
• Application Number:
  US07/524,663
• Time in Patent Office: 504 Days
• Field of Search: 381/41-46, 364/513.5
• US Class Current: 704/254
• CPC Class Codes: G10L 15/10 using distance or distortion measures between unknown speech and reference templates
• CPC Class Codes: G10L 2015/027 Syllables being the recognition units

Abstract

Syllables are recognized in voice data obtained from an input voice signal. Character arrays are used to represent the types and time frames of syllables with which the voice data are compared.

Claims

1. A method of recognizing syllables in an input voice signal comprising the steps of

sequentially sampling an input voice signal at regular time intervals to obtain a series of sampled signals,

comparing said sampled signals with standard phoneme patterns to phonetically characterize each of said sampled signals and assigning character symbols individually to said sampled signals to thereby obtain an array of said character symbols, said character symbols being each indicative of a phonological characteristic,

obtaining an array of segments by partitioning said array of character symbols into segments, each of said segments containing at least one of said character symbols and having its length defined by the number of said character symbols contained therein,

providing a collection of standard syllable patterns, each of said standard syllable patterns being identified as representing a syllable and comprising an array of character groups, each of said character groups including one of said character symbols and a range symbol indicating a range of duration, said range being specified as being between a minimum and a maximum, said minimum and maximum being selected from more than three predefined values, and

assigning syllables to said array of segments by sequentially comparing said array of segments with both said character and range symbols of said standard syllable patterns.

2. The method of claim 1 wherein said range symbol includes a smaller one-digit integer n indicative of said minimum and a larger one-digit integer N indicative of said maximum.

3. The method of claim 2 wherein said minimum is given as a function of n as (n² +n+2)/2 and said maximum is given as a function of N as (N²⁺ 3N+2)/2.

4. The method of claim 1 wherein said character symbols include consonant symbols each representing a consonant and vowel symbols each representing a vowel.

5. An apparatus for recognizing syllables in an input voice signal comprising

sampling means for sequentially sampling an input voice signal at regular time intervals to obtain a series of sampled signals,

means for comparing said sampled signals with standard phoneme patterns to phonetically characterize each of said sampled signals and assigning character symbols individually to said sampled signals to thereby obtain an array of said character symbols, said character symbols being each indicative of a phonological characteristic,

means for obtaining an array of segments by partitioning said array of character symbols into segments, each of said segments containing at least ne of said character symbols and having its length defined by the number of said character symbols contained therein,

dictionary means containing a collection of standard syllable patterns, each of said standard syllable patterns being identified as representing a syllable and comprising an array of character groups, each of said character groups including one of said character symbols and a range symbol indicating a range of duration, said range being specified as being between a minimum and a maximum, said minimum and maximum being selected from more than three predefined values, and

processing means for assigning syllables to said array of segments by sequentially comparing said array of segments with both said character and range symbols of said standard syllable patterns.

6. The apparatus of claim 5 wherein said range symbol includes a smaller one-digit integer n indicative of said minimum and a larger one-digit integer N indicative of said maximum.

7. The apparatus of claim 6 wherein said minimum is given as a function of n as (n² +n+2)/2 and said maximum is given as a function of N as (N² +3N+2)/2.

8. The apparatus of claim 5 wherein said character symbols include consonant symbols each representing a consonant and vowel symbols each representing a vowel.