Speech information processing method and apparatus and storage medium using a segment pitch pattern model
First Claim
Patent Images
1. A speech information processing method comprising:
- an input step of inputting speech;
an extraction step of extracting a feature parameter of the speech; and
a speech recognition step of recognizing the feature parameter based on a segment pitch pattern model,wherein the segment pitch pattern model is obtained by modeling time change in a fundamental frequency of a phoneme belonging to a predetermined phonemic environment with a polynomial segment model.
0 Assignments
0 Petitions
Accused Products
Abstract
A speech information processing apparatus and method performs speech recognition. Speech is input, and feature parameters of the input speech are extracted. The feature parameters are recognized based on a segment pitch pattern model. The segment pitch pattern model may be obtained by modeling time change in a fundamental frequency of a phoneme belonging to a predetermined phonemic environment with a polynomial segment model. The segment pitch pattern model may also be obtained by modeling with at least one of a single mixed distribution and a multiple mixed distribution.
53 Citations
6 Claims
-
1. A speech information processing method comprising:
-
an input step of inputting speech; an extraction step of extracting a feature parameter of the speech; and a speech recognition step of recognizing the feature parameter based on a segment pitch pattern model, wherein the segment pitch pattern model is obtained by modeling time change in a fundamental frequency of a phoneme belonging to a predetermined phonemic environment with a polynomial segment model. - View Dependent Claims (3)
-
-
2. A speech information processing method comprising:
-
an input step of inputting speech; an extraction step of extracting a feature parameter of the speech; and a speech recognition step of recognizing the feature parameter based on a segment pitch pattern model, wherein the segment pitch pattern model is obtained by modeling with at least one of a single mixed distribution and a multiple mixed distribution. - View Dependent Claims (6)
-
-
4. A speech information processing apparatus comprising:
-
input means for inputting speech; extraction means for extracting a feature parameter of the speech; and speech recognition means for recognizing the feature parameter based on a segment pitch pattern model, wherein the segment pitch pattern model is obtained by modeling time change in a fundamental frequency of a phoneme belonging to a predetermined phonemic environment with a polynomial segment model.
-
-
5. A speech information processing apparatus comprising:
-
input means for inputting speech; extraction means for extracting a feature parameter of the speech; and speech recognition means for recognizing the feature parameter based on a segment pitch pattern model, wherein the segment pitch pattern model is obtained by modeling with at least one of a single mixed distribution and a multiple mixed distribution.
-
Specification