Continuous speech recognition apparatus

Claim 2

A continuous speech recognition apparatus comprising: input pattern output means for outputting a signal corresponding to continuously uttered speech as an input pattern; reference pattern output means for outputting a plurality of connected reference patterns obtained by connecting previously prepared reference patterns; start vicinity setting means for setting a start vicinity of said input pattern; first dissimilarity measure calculating means for calculating a first dissimilarity measure between said input pattern and said reference pattern, proportional to the time length of said reference pattern in a portion for a head word of said input pattern; and second dissimilarity measure calculating means for calculating a second dissimilarity measure proportional to the time length of said input pattern from a temporary start point on the basis of said first dissimilarity measure, said temporary start point ranging within said start vicinity.

Claim 3

A continuously speech recognition apparatus according to claim 1, further comprising a third dissimilarity measure calculating means for calculating a third dissimilarity measure between said input pattern and said reference pattern on second and following digits, which is proportional to the time length of said input pattern based upon said second dissimilarity measure as a boundary condition, said digit being the connection order number of said connected reference pattern.

Claim 4

A continuous speech recognition apparatus according to claim 2, further comprising an end vicinity setting means for determining the end vicinity of said input pattern and an end decision means for deciding an end point by using a minimum normalized dissimilarity measure normalized by the time length of said input pattern in said determined end vicinity.

Claim 5

A continuous speech recognition apparatus according to claim 3, further comprising a recognition means for outputting said connected reference pattern giving minimum normalized dissimilarity measure as a recognition result.

Claim 6

A continuous speech recognition apparatus according to claim 1, wherein said start vicinity setting means is a means for setting as said start vicinity a portion having a given time length before and after a temporary start point at which the level of said uttered speech becomes greater than a given value.

Claim 7

A continuous speech recognition apparatus according to claim 3, wherein said end vicinity setting means is a means for setting a portion having a given time length before and after the time point at which the level of said uttered speech becomes smaller than a given value at said vicinity of the end point.

Claim 8

A continuous speech recognition apparatus according to claim 1, wherein said conversion in said second dissimilarity measure calculating means is performed in accordance with the following equation: G(i)=(i/J.sup.n)·g(i,J.sup.n) where G(i) denotes the second dissimilarity measure; J n , the time length of the reference pattern; i, the time length of the input pattern; amd g(i,J n ), the first dissimilarity measure obtained at the end time of a reference pattern n.

Claim 9

A continuous speech recognition apparatus comprising: input pattern output means for outputting a signal corresponding to continuously uttered speech as an input pattern; reference pattern output means for outputting a connected pattern of the reference patterns specified by a definite-state automation; start point matching means for determining a disimilarity measure proportional to the time length of the reference pattern between said reference pattern in the initial state of the definite-state automation and a portion of a head word of said input pattern; dissimilarity measure converting means for converting the dissimilarity measure obtained by said start point matching means into a dissimilarity measure proportional to the time length of said input pattern from a temporary start point, said temporary start point ranging within a vicinity of said input pattern; main matching means for determining a dissimilarity measure proportional to the time length of said input pattern between said input pattern and said reference pattern in each state other than said initial state of said definite-state automation in which said converted dissimilarity measure is used as a boundary condition; end point decision means for determining an end point having a minimum normalized dissimilarity measure normalized by the time length of said input pattern in an end vicinity; and a recognition means for obtaining recognition results by determining the sequence of said reference pattern along a dynamic programming-matching pass, through which said minimum normalized dissimilarity measure is obtained.

Claim 10

A continuous speech recognition apparatus comprising: input pattern output means for outputting a signal corresponding to continuously uttered speech as an input pattern; reference pattern output means for outputting a connected pattern of the reference patterns specified by a definite-state automation; start point matching means for determining a disimilarity measure proportional to the time length of the reference pattern between said reference pattern in the initial state of the definite-state automation and a portion of a head word of said input pattern; dissimilarity measure converting means for converting the dissimilarity measure obtained by said start point matching means into a dissimilarity measure proportional to the time length of said input pattern from a temporary start point, said temporary start point ranging within a vicinity of said input pattern; main matching means for determining a dissimilarity measure proportional to the time length of said input pattern between said input pattern and said reference pattern in each state other than said initial state of said definite-state automation in which said converted dissimilarity measure is used as a boundary condition; end point decision means for determining an end point having a minimum normalized dissimilarity measure normalized by the time length of said input pattern in an end vicinity; and a recognition means for obtaining recognition results by determining the sequence of said reference pattern along a dynamic programming-matching pass, through which said minimum normalized dissimilarity measure is obtained.

Claim 11

A continuous speech recognition apparatus comprising: means for generating a signal corresponding to continuously uttered speech as an input pattern; reference pattern output means for outputting a plurality of connected reference patterns specified by a definite-state atomation; start point matching means for determining a dissimilarity measure proportional to the time length of the reference pattern between said reference pattern in the initial state of the definite-state automation and a proportion of a head word of said input pattern; dissimilarity measure converting means for converting the dissimilarity measure obtained by said start point matching means into a dissimilarity measure proportional to the time length of said input pattern from a temporary start point, said temporary start point ranging within a vicinity of said input pattern; and main matching means for determining a dissimilarity measure proportional to the time length of said input pattern between said input pattern and said reference pattern in each state other than said initial state of said defininte-state automation in which said converted dissimilarity measure is used as a boundary condition.

Claim 12

A continuous speech recognition apparatus comprising: means for generating a signal corresponding to continuously uttered speech as an input pattern; reference pattern output means for outputting a plurality of connected reference patterns specified by a definite-state atomation; start point matching means for determining a dissimilarity measure proportional to the time length of the reference pattern between said reference pattern in the initial state of the definite-state automation and a proportion of a head word of said input pattern; dissimilarity measure converting means for converting the dissimilarity measure obtained by said start point matching means into a dissimilarity measure proportional to the time length of said input pattern from a temporary start point, said temporary start point ranging within a vicinity of said input pattern; and main matching means for determining a dissimilarity measure proportional to the time length of said input pattern between said input pattern and said reference pattern in each state other than said initial state of said defininte-state automation in which said converted dissimilarity measure is used as a boundary condition.

Claim 13

A continuous speech recognition apparatus according to claim 9, further comprising: end point decision means for determining an end point having a minimum normalized dissimilarity measure normalized by the time length of said input pattern in an end vicinity.

Claim 14

A continuous speech recognition apparatus according to claim 10, further comprising recognition means for obtaining recognition results by determining the sequence of said reference pattern along a dynamic programming matching pass, through which said minimum normalized dissimilarity measure is obtained.