Method and device of matching speech input to text

• US 9,437,192 B2
• Filed: 01/24/2014
• Issued: 09/06/2016
• Est. Priority Date: 06/21/2013
• Status: Active Grant

• Current Assignee: Tencent Technology Shenzhen Company Limited (Tencent Holdings Limited)
• Original Assignee: Tencent Technology Shenzhen Company Limited (Tencent Holdings Limited)
• Inventors: Han, Guiping, Xu, Hao, Zhu, Pan, Wang, Fei, Liao, Ling
• Primary Examiner(s): Hudspeth, David
• Assistant Examiner(s): Nguyen, Timothy
• Application Number:
  US14/164,075
• Publication Number:
  US 20140379335A1
• Time in Patent Office: 956 Days
• Field of Search: 704/235
• US Class Current: 1/1
• CPC Class Codes: G06F 40/129 Handling non-Latin characters, e.g. kana-to-kanji conversion
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A method and device for matching speech to text are disclosed, the method including: receiving a speech input, the mentioned speech input carrying input speech information; obtaining initial text corresponding to the input speech information, and respective pinyin of the initial text; generating at least one approximate pinyin for the initial text based on predetermined pronunciation similarity information; and from a preset mapping relationship table, obtaining additional text corresponding to the respective pinyin of the initial text or to the at least one approximate pinyin of the initial text, wherein the preset mapping relationship table includes a respective record for each word in a word database, including respective pinyin and at least one respective approximate pinyin for said each word, and a respective mapping relation between said respective pinyin, said at least one respective approximate pinyin, and said each word.

Claims

1. A method of matching speech to a contact name in a user'"'"'s list of contacts, comprising:

at a device having one or more processors and memory;

storing the list of the user'"'"'s contacts in the memory, wherein the list of the user'"'"'s contacts includes a first contact;

receiving input speech information;

obtaining initial text corresponding to the input speech information and standard pinyin of the initial text;

generating at least one approximate pinyin for the initial text based on predetermined pronunciation similarity information, wherein the pronunciation similarity information identifies non-standard pronunciations based on regional dialects;

from a preset mapping relationship table stored in the memory, obtaining alternative text corresponding to the approximate pinyin of the initial text, wherein;

the preset mapping relationship table includes a record for each word in a word database, the record including standard pinyin and approximate pinyin for the word, and a mapping relation between the standard pinyin, approximate pinyin, and the word; and

the alternative text includes one or more words whose standard pinyin or approximate pinyin map, in the preset mapping relationship table, to the approximate pinyin of the initial text;

matching the alternative text to the first contact; and

performing an operation using the first contact.

2. The method of claim 1, further comprising:

establishing the preset mapping relationship table, comprising, for each word in the word database;

converting the word to corresponding standard pinyin;

using the standard pinyin, generating at least one approximate pinyin according to the predetermined pronunciation similarity;

establishing the mapping relationship between the standard pinyin, the approximate pinyin, and the word; and

storing the respective mapping relationship for the word in the preset mapping relationship table.

3. The method of claim 2, wherein establishing the preset mapping relationship table further includes establishing a dictionary Trie for the standard pinyin and the at least one approximate pinyin corresponding to each word in the word database.

4. The method of claim 2, wherein the words in the word database are divided into a plurality of categories and a respective dictionary Trie is established for each of the plurality of categories.

5. The method of claim 1, wherein the preset mapping relationship table includes respective approximate pinyin of different approximation grades corresponding to each word in the word database.

6. The method of claim 5, wherein obtaining the alternative text corresponding to the approximate pinyin of the initial text further comprises:

according to a descending order of the different approximation grades, successively determining whether the standard pinyin of the initial text matches pinyin in the preset mapping relationship table; and

in accordance with a determination that the standard pinyin of the initial text matches pinyin in the preset main relationship table, obtaining, as the alternative text, text corresponding to the one or more matching pinyin identified according to the preset mapping relationship table.

7. A system, comprising:

one or more processors; and

memory storing instructions which, when executed by the one or more processors, cause the processors to perform operations comprising;

storing a list of a user'"'"'s contacts in the memory, wherein the list of the user'"'"'s contacts includes a first contact;

receiving input speech information;

obtaining initial text corresponding to the input speech information and standard pinyin of the initial text;

generating at least one approximate pinyin for the initial text based on predetermined pronunciation similarity information, wherein the pronunciation similarity information identifies non-standard pronunciations based on regional dialects;

from a preset mapping relationship table stored in the memory, obtaining alternative text corresponding to the approximate pinyin of the initial text, wherein;

the preset mapping relationship table includes a record for each word in a word database, the record including standard pinyin and approximate pinyin for the word, and a mapping relation between the standard pinyin, approximate pinyin, and the word; and

the alternative text includes one or more words whose standard pinyin or approximate pinyin map, in the preset mapping relationship table, to the approximate pinyin of the initial text;

matching the alternative text to the first contact; and

performing an operation using the first contact.

8. The system of claim 7, wherein the operations further comprise:

establishing the preset mapping relationship table, comprising, for each word in the word database;

converting the word to corresponding standard pinyin;

using the standard pinyin, generating at least one approximate pinyin according to the predetermined pronunciation similarity;

establishing the mapping relationship between the standard pinyin, the approximate pinyin, and the word; and

storing the respective mapping relationship for the word in the preset mapping relationship table.

9. The system of claim 8, wherein establishing the preset mapping relationship table further includes establishing a dictionary Trie for the standard pinyin and the at least one approximate pinyin corresponding to each word in the word database.

10. The system of claim 8, wherein the words in the word database are divided into a plurality of categories and a respective dictionary Trie is established for each of the plurality of categories.

11. The system of claim 7, wherein the preset mapping relationship table includes respective approximate pinyin of different approximation grades corresponding to each word in the word database.

12. The system of claim 11, wherein obtaining the alternative text corresponding to the approximate pinyin of the initial text further comprises:

according to a descending order of the different approximation grades, successively determining whether the standard pinyin of the initial text matches pinyin in the preset mapping relationship table; and

in accordance with a determination that the standard pinyin of the initial text matches pinyin in the preset main relationship table, obtaining, as the alternative text, text corresponding to the one or more matching pinyin identified according to the preset mapping relationship table.

13. A non-transitory computer-readable storage medium having storing instructions which, when executed by one or more processors, cause the processors to perform operations comprising:

storing a list of a user'"'"'s contacts in the memory, wherein the list of the user'"'"'s contacts includes a first contact;

receiving input speech information;

obtaining initial text corresponding to the input speech information and standard pinyin of the initial text;

generating at least one approximate pinyin for the initial text based on predetermined pronunciation similarity information, wherein the pronunciation similarity information identifies non-standard pronunciations based on regional dialects;

from a preset mapping relationship table stored in the memory, obtaining alternative text corresponding to the approximate pinyin of the initial text, wherein;

the preset mapping relationship table includes a record for each word in a word database, the record including standard pinyin and approximate pinyin for the word, and a mapping relation between the standard pinyin, approximate pinyin, and the word; and

the alternative text includes one or more words whose standard pinyin or approximate pinyin map, in the preset mapping relationship table, to the approximate pinyin of the initial text;

matching the alternative text to the first contact; and

performing an operation using the first contact.

14. The computer-readable medium of claim 13, wherein the operations further comprise:

establishing the preset mapping relationship table, comprising, for each word in the word database;

converting the word to corresponding standard pinyin;

using the standard pinyin, generating at least one approximate pinyin according to the predetermined pronunciation similarity;

establishing the mapping relationship between the standard pinyin, the approximate pinyin, and the word; and

storing the respective mapping relationship for the word in the preset mapping relationship table.

15. The computer-readable medium of claim 14, wherein establishing the preset mapping relationship table further includes establishing a dictionary Trie for the standard pinyin and the at least one approximate pinyin corresponding to each word in the word database.

16. The computer-readable medium of claim 14, wherein the words in the word database are divided into a plurality of categories and a respective dictionary Trie is established for each of the plurality of categories.

17. The computer-readable medium of claim 13, wherein the preset mapping relationship table includes respective approximate pinyin of different approximation grades corresponding to each word in the word database.

18. The computer-readable medium of claim 17, wherein obtaining the alternative text corresponding to the approximate pinyin of the initial text further comprises:

according to a descending order of the different approximation grades, successively determining whether the standard pinyin of the initial text matches pinyin in the preset mapping relationship table; and

in accordance with a determination that the standard pinyin of the initial text matches pinyin in the preset mapping relationship table, obtaining, as the alternative text, text corresponding to the one or more matching pinyin identified according to the preset mapping relationship table.