Method for text-to-pronunciation conversion capable of increasing the accuracy by re-scoring graphemes likely to be tagged erroneously

• US 20050197838A1
• Filed: 07/28/2004
• Published: 09/08/2005
• Est. Priority Date: 03/05/2004
• Status: Abandoned Application

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Lin, Yi-Chung, Wang, Ren-Jr, Hung, Peng-Hsiang
• Application Number:
  US10/900,101
• Publication Number:
  US 20050197838A1
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams

Abstract

The present invention provides a method for text-to-pronunciation conversion capable of increasing the accuracy by re-scoring graphemes likely to be tagged erroneously. Grapheme segmentation and phoneme tagging are first applied to an input word to generate at least one grapheme-phoneme pair sequence, and the score of each grapheme-phoneme pair sequence is also computed. Then, at least one grapheme-phoneme pair sequence having a higher score is selected. For the selected grapheme-phoneme pair sequence that has a grapheme likely to be tagged erroneously, the features in the context of the grapheme are selected and made good use of computing re-score corresponding to the graphemes likely to be tagged erroneously, so as to re-score the grapheme-phoneme pair sequence. Accordingly, the grapheme-phoneme pair sequence with the highest score is the final conversion result.

Claims

1. A method for text-to-pronunciation conversion capable of increasing the accuracy by re-scoring graphemes likely to be tagged erroneously, the method comprising:

applying grapheme segmentation and phoneme tagging to an input word to generate at least one grapheme-phoneme pair sequence, every grapheme-phoneme pair sequence comprising at least one grapheme and a corresponding phoneme, and computing a score of each grapheme-phoneme pair sequence; and

re-scoring a grapheme-phoneme pair sequence that has a grapheme likely to be tagged erroneously from the at least one grapheme-phoneme pair sequence having a higher score, features in a context of the grapheme being selected and utilized for computing a connection between the features and phoneme corresponding to the grapheme likely to be tagged erroneously thereby re-scoring the grapheme-phoneme pair sequence, and accordingly, using the grapheme-phoneme pair sequence with the highest score as a final conversion result.

2. The method as claimed in claim 1, wherein the connection between the features and phoneme corresponding to the grapheme likely to be tagged erroneously is computed by mutual information.

3. The method as claimed in claim 1, wherein the generation of the grapheme-phoneme pair sequence comprises:

applying the grapheme segmentation to the input word according to graphemes stored in a predetermined grapheme set to obtain at least one grapheme sequence and a corresponding score, every grapheme sequence comprising a plurality of graphemes;

applying phoneme tagging to at least one grapheme sequence having a higher score according to a predetermined mapping between the grapheme and the phoneme to obtain at least one phoneme sequence for every grapheme sequence and a score for every phoneme sequence, and then selecting at least one phoneme sequence having a higher score to generate at least one grapheme-phoneme pair sequence.

4. The method as claimed in claim 2, wherein every grapheme-phoneme pair sequence is re-scored to have a score as:

$S_R=\underset{g_i\in <br><br>E}{\underset{i}{\sum <br><br>}}<br><br>\underset{j=1}{\overset{<br><br>X<br><br>\left(i\right)<br><br>}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>w_j<br><br>\log <br><br>\left(\frac{P<br><br>\left(x_j,f_i\right)}{P<br><br>\left(x_j\right)<br><br>P<br><br>\left(f_i\right)}\right)\times <br><br>\frac{1}{\underset{g_i\in <br><br>E}{\underset{i}{\sum <br><br>}}<br><br>1}$

where g_i is a grapheme of the grapheme sequence, f_i is a phoneme of the phoneme sequence, w_j is a weight value, E is a set of graphemes likely to be tagged erroneously, X(i) is a set of selected features, x_j represents any one feature in the feature set X(i).

5. The method as claimed in claim 4, wherein X(i) is:

$\begin{array}{c}X<br><br>\left(i\right)=\stackrel{N}{\underset{n=1}{\bigcup <br><br>}}<br><br>X_n<br><br>\left(i;<br><br>g\right)⋓<br><br>\stackrel{N}{\underset{n=1}{\bigcup <br><br>}}<br><br>X_n<br><br>\left(i;<br><br>f\right)⋓<br><br>\stackrel{N}{\underset{n=1}{\bigcup <br><br>}}<br><br>X_n<br><br>\left(i;<br><br>\mathrm{\tau <br><br>}\right)\\ X_n<br><br>\left(i;<br><br>y\right)=\{x|x=y_l<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>y_r,i-L\le <br><br>l\le <br><br>r\le <br><br>i+R\bigwedge <br><br>\left(r-l+1\right)\\ \text{\hspace{1em}<br><br>}<br><br>=n\bigwedge <br><br>i\notin <br><br>\left[l,r\right]\}<br><br>\text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}<br><br>⋓<br><br>\{x|x=y_l<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>y_{i-1}<br><br>y_{i+1}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>y_r,\\ \text{\hspace{1em}<br><br>}<br><br>i-L\le <br><br>l\le <br><br>r\le <br><br>i+R\bigwedge <br><br>\left(r-l+1\right)\\ \text{\hspace{1em}<br><br>}<br><br>=n\bigwedge <br><br>i\in <br><br>\left[l,r\right]\}\end{array}$

where τ

_i≡

g_if_i, L and R represent a range of a context of the grapheme g_i, N is a number of selected grapheme-phoneme pair sequences having higher scores, y is g, f or τ

, and l and r represent the position of y that needs to be between i−

L and i+R.

6. The method as claimed in claim 3 wherein a score S_G2P of every grapheme-phoneme pair sequence is:

S_G2P=w_GS_G+w_PS_P,

where S_G is a score of the grapheme sequence, S_P is a score of the phoneme sequence, and W_G and W_P are weight values.

7. The method as claimed in claim 6, wherein in the grapheme segmentation, an obtained score S_G of every grapheme sequence is:

$S_G=\underset{i=1}{\overset{n}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>\log <br><br>\left(P<br><br>\left(g_i|g_{i-N+1}^{i-1}\right)\right),$

where g_i is a grapheme of the grapheme sequence, n is a number of graphemes included in the grapheme sequence, N is a score of g_i decided by N graphemes before g_i.

8. The method as claimed in claim 6, wherein in the phoneme tagging, an obtained score S_P of every phoneme sequence is:

$S_P=\underset{i=1}{\overset{n}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>\log <br><br>\left(P<br><br>\left(f_i|g_{i-R}^{i-L}\right)\right),$

where f_i is a phoneme of the phoneme sequence, L and R represent two ranges of a context of the grapheme g_i, and n is a number of phonemes included in the phoneme sequence.

9. The method as claimed in claim 4, wherein in re-scoring, a re-scored score S_Final of every grapheme-phoneme pair sequence is:

S_Final=w_G2PS_G2P+w_RS_R,

where W_G2P and W_R are weight values.

10. The method as claimed in claim 1, wherein the input word is Romanic text.

11. The method as claimed in claim 1, wherein the graphemes likely to be tagged erroneously are vowels in English.

12. The method as claimed in claim 1, wherein the features in the context include phoneme, grapheme and grapheme-phoneme pair.

13. The method as claimed in claim 3, wherein in the phoneme tagging, every grapheme corresponds to at least one phoneme.

14. The method as claimed in claim 3, wherein in the grapheme segmentation, an N-gram module is used to perform the grapheme segmentation to the input text.