Method to synthesize personalized phonetic transcription

Patent metadata

Field	Value
Publication number	US-9990916-B2
Application number	US-201615139141-A
Country	US
Kind code	B2
Filing date	Apr 26, 2016
Priority date	Apr 26, 2016
Publication date	Jun 5, 2018
Grant date	Jun 5, 2018

Abstract

Technology related to improving synthesis of foreign regional nouns using personalized and culturally correct phonetic transcription is described. The technology includes systems and methods for generating personalized speech by receiving an input, the input including textual data, identifying a regional noun in the textual data, and determining a user accent classification based on a context of the input. The method may further include determining a personalized phonetic transcription of the regional noun corresponding to the user accent classification and using a phonetic inventory stored in a database and outputting the personalized phonetic transcription.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving an input, the input including textual data;
identifying a regional noun in the textual data;
determining a user accent classification based on a context of the input;
accessing a phonetic inventory stored in a database, in which multiple phonetic transcriptions are stored for each of a plurality of regional nouns, including the regional noun;
executing a search of the database using the user accent classification and the regional noun as search parameters;
determining a personalized phonetic transcription of the regional noun, selected from the multiple phonetic transcriptions stored for the regional noun as a search result of executing the search of the phonetic inventory stored in the database;
outputting the personalized phonetic transcription; and
synthesizing audio output of the textual data, based on the personalized phonetic transcription.

2. The computer-implemented method of claim 1 , wherein determining the personalized phonetic transcription of the regional noun is based on speech patterns of a user.

3. The computer-implemented method of claim 1 , wherein identifying a regional noun in the textual data includes narrowing a set of regional nouns searched based on a geographic location in which a user device is located, the user device being a device via which the input is received.

4. The computer-implemented method of claim 1 , wherein the context of the input includes a vernacular of a user.

5. The computer-implemented method of claim 4 , further comprising determining the vernacular of the user based on an online presence of the user.

6. The computer-implemented method of claim 1 , wherein the textual data includes orthographic text.

7. A computer-implemented method comprising:
receiving an audio input, the audio input including speech;
identifying a regional noun in the speech;
generating a phonetic transcription of the regional noun using the audio input;
determining a user accent classification based on a context of the audio input;
executing a search of a phonetic inventory stored in a database, using the generated phonetic transcription and the user accent classification as search parameters, wherein the phonetic inventory stores multiple phonetic transcriptions for each of a plurality of regional nouns, including the regional noun;
returning the stored phonetic transcription of the regional noun as a search result of executing the search of the phonetic inventory stored in the database;
translating the regional noun into textual data using the stored phonetic transcription; and
outputting the textual data.

8. The computer-implemented method of claim 7 , wherein generating the phonetic transcription of the regional noun includes personalizing the phonetic transcription for a user based on speech patterns of the user.

9. The computer-implemented method of claim 7 , wherein identifying a regional noun in the textual data includes narrowing a set of regional nouns searched based on a geographic location in which a user device is located, wherein the user device is a device via which the audio input is received.

10. The computer-implemented method of claim 7 , wherein the textual data includes orthographic text.

11. The computer-implemented method of claim 7 , wherein the context of the input further includes a vernacular of a user.

12. The computer-implemented method of claim 11 , further comprising determining the vernacular based on an online presence of the user.

13. The computer-implemented method of claim 11 , wherein outputting the textual data includes outputting a localized spelling of orthographic text based on the vernacular of the user.

14. A system comprising:
one or more processors; and
a memory coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to:
receive a textual input including a word;
determine a user accent classification based on a context of the textual input;
accessing a phonetic inventory stored in a database, in which multiple phonetic transcriptions are stored for each of a plurality of words, including the word;
executing a search of the database using the user accent classification and the regional noun as search parameters;
determine a personalized phonetic transcription of the word, selected from the multiple phonetic transcriptions stored for the word as a search result of executing the search of the phonetic inventory stored in the database;
output the personalized phonetic transcription; and
synthesize audio output of the textual input, based on the personalized phonetic transcription.

15. The system of claim 14 , wherein to determine the personalized phonetic transcription of the word is based on speech patterns of a user.

16. The system of claim 14 , wherein the instructions, when executed by the one or more processors, further cause the system to narrow a set of words in the phonetic inventory against which the personalized phonetic transcription is determined using a geographic location in which a user device is located, wherein the user device is a device via which the textual input is received.

17. The system of claim 14 , wherein the context of the textual input includes a vernacular of a user.

18. The system of claim 17 , wherein the instructions, when executed by the one or more processors, further cause the system to determine the vernacular of the user based on a current or past residence of the user.

19. The system of claim 14 , wherein the textual input includes orthographic text.

Assignees

• Adobe Systems Inc

Inventors

• Vanreusel Jean-Francois
• Kumar Vikas
• Jha Nandan

Classifications

• G06F40/30

  Semantic analysis · CPC title

• G06F40/263

  Language identification · CPC title

• G06F40/242

  Dictionaries · CPC title

• G10L15/07 Primary

  to the speaker · CPC title

• G10L13/08 Primary

  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 title