Voice labeling error detecting system, voice labeling error detecting method and program

• US 20050060144A1
• Filed: 08/18/2004
• Published: 03/17/2005
• Est. Priority Date: 08/27/2003
• Status: Active Grant

• Current Assignee: Rakuten Group, Inc.
• Original Assignee: Kabushiki Kaisha Kenwood
• Inventors: Koyama, Rika
• Granted Patent: US 7,454,347 B2
• Time in Patent Office: Days
• US Class Current: 704/205
• CPC Class Codes: G10L 13/06 Elementary speech units used in speech synthesisers; Concatenation rules

Abstract

A labeling part 3 analyzes the character string data to produce a phoneme label and a prosody label, partition the voice data stored in a voice database 1 into phonemic data, and label the phonemic data, employing the phoneme label and the like. A phoneme segmenting part 4 connects the voice data labeled with the same kind of phonemic data, and a formant extracting part 5 specifies the frequency of formant of each piece of phonemic data. A processing part 6 decides an evaluation value for each phonemic data based on the frequency of formant, and an error detection part 7 detects the phonemic data of which a deviation of the evaluation value within a set of phonemic data reaches a predetermined amount.

Claims

1. A voice labeling error detecting system comprising:

data acquisition means for acquiring the waveform data representing a waveform of a unit voice and the labeling data for identifying the kind of said unit voice;

classification means for classifying the waveform data acquired by said data acquisition means into the kinds of unit voice, based on the labeling data acquired by said data acquisition means;

evaluation value decision means for specifying a frequency of a formant of each unit voice represented by the waveform data acquired by said data acquisition means and determining an evaluation value of said waveform data based on the specified frequency; and

error detection means for detecting the waveform data from among a set of waveform data classified into the same kind, for which a deviation of evaluation value within said set reaches a predetermined amount, and outputting the data representing said detected waveform data, as waveform data having a labeling error.

2. The voice labeling error detecting system according to claim 1, characterized in that said evaluation value is a linear combination of the values {|f(k)−

F(k)|} where the value of k is an integer from 1 to n, assuming that F(k) is the frequency of the k-th formant of a unit voice indicated by the waveform data to calculate the evaluation value, and f(k) is the average value of the frequency of the k-th formant of the unit voice indicated by each waveform data classified into the same kind as said waveform data.

3. The voice labeling error detecting system according to claim 1, characterized in that said evaluation value is a linear combination of plural frequencies of formants in the spectrum of acquired waveform data.

4. The voice labeling error detecting system according to claim 1, 2, or 3, characterized in that said evaluation value deciding means deals with the frequency at the maximal value of the spectrum in the waveform data as the frequency of formant of unit voice indicated by said waveform data.

5. The voice labeling error detecting system according to any one of claims 1 to 3, characterized in that said evaluation value deciding means specifies the order of formant used to decide the evaluation value of the waveform data as the kind of unit voice indicated by said waveform data, corresponding to the kind of labeling data.

6. The voice labeling error detecting system according to any one of claims 1 to 3, characterized in that said error detection means detects the waveform data associated with the labeling data indicating a voiceless state at which the magnitude of voice represented by said waveform data reaches a predetermined amount as the waveform data in which the labeling has an error.

7. The voice labeling error detecting system according to any one of claims 1 to 3, characterized in that said classification means comprises means for concatenating each waveform data classified into the same kind in the form in which two adjacent pieces of waveform data sandwiches data indicating the voiceless state therebetween.

8. A voice labeling error detecting method comprising the steps of:

acquiring the waveform data representing a waveform of a unit voice and the labeling data for identifying the kind of said unit voice;

classifying said acquired waveform data into the kinds of unit voice, based on said acquired labeling data;

specifying a frequency of a formant of each unit voice represented by the waveform data and deciding an evaluation value of said waveform data based on the specified frequency; and

detecting the waveform data having a labeling error, from among a set of waveform data classified into the same kind, in which a deviation of evaluation value within said set reaches a predetermined amount and outputting data representing said detected waveform data.

9. A program for enabling a computer to operate as:

data acquisition means for acquiring the waveform data representing a waveform of a unit voice and the labeling data for identifying the kind of said unit voice;

classification means for classifying the waveform data acquired by said data acquisition means into the kinds of unit voice, based on the labeling data acquired by said data acquisition means;

evaluation value decision means for specifying a frequency of a formant of each unit voice represented by the waveform data acquired by said data acquisition means and deciding an evaluation value of said waveform data based on the specified frequency; and

error detection means for detecting the waveform data having a labeling error, from among a set of waveform data classified into the same kind, in which a deviation of evaluation value within said set reaches a predetermined amount, and outputting the data representing said detected waveform data.