METHOD AND APPARATUS FOR DISTRIBUTED VOICE SEARCHING

• US 20080256033A1
• Filed: 04/10/2007
• Published: 10/16/2008
• Est. Priority Date: 04/10/2007
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: CHENG, Yan Ming
• Granted Patent: US 7,818,170 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/433 using audio data
• CPC Class Codes: G06F 16/957 Browsing optimisation, e.g. caching or content distillation
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: H04M 1/72445 for supporting Internet browser applications
• CPC Class Codes: H04M 2250/74 with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00

Abstract

A method for distributed voice searching may include receiving a search query from a user of the mobile communication device, generating a lattice of coarse linguistic representations from speech parts in the search query, extracting query features from the generated lattice of coarse linguistic representations, generating coarse search feature vectors based on the extracted query features, performing a coarse search using the generated coarse search feature vectors and transmitting the generated coarse search feature vectors to a remote voice search processing unit, receiving remote resultant web indices from the remote voice search processing unit, generating a lattice of fine linguistic representations from speech parts in the search query, generating fine search feature vectors from the lattice of fine linguistic representations, performing a fine search using the coarse search results, the remote resultant web indices and the generated fine search feature vectors, and displaying the fine search results to the user.

Claims

1. A method for distributed voice searching using a mobile communication device, comprising:

receiving a search query from a user of the mobile communication device;

generating a lattice of coarse linguistic representations from speech parts in the search query;

extracting query features from the generated lattice of coarse linguistic representations;

generating coarse search feature vectors based on the extracted query features;

performing a coarse search using the generated coarse search feature vectors and transmitting the generated coarse search feature vectors to a remote voice search processing unit;

receiving remote resultant web indices from the remote voice search processing unit based on the generated coarse search feature vectors;

generating a lattice of fine linguistic representations from speech parts in the search query based on the coarse search results and the remote resultant web indices;

generating fine search feature vectors from the lattice of fine linguistic representations;

performing a fine search using the coarse search results, the remote resultant web indices and the generated fine search feature vectors; and

displaying the fine search results to the user.

2. The method of claim 1, wherein the coarse linguistic representations and the fine linguistic representations are at least one of words, phrases, morphemes, syllables, and phonemes, of at least one language.

3. The method of claim 1, wherein the coarse search feature vectors and the fine search feature vectors are generated using linguistic metadata from a metadata database, the metadata database containing multimedia content, the multimedia content being at least one of audio files, audio recordings, emails, voice mails, recorded conversations, notes, messages, text messages, photos, video files, movie files, and television recordings.

4. The method of claim 1, wherein the remote resultant web indices result from at least one search conducted by at least one web search engine.

5. The method of claim 1, wherein the coarse search results and the remote resultant web indices contain a plurality of candidate selections based on the coarse search feature vectors.

6. The method of claim 1, wherein the fine search results contain at least one best candidate from a combination of the coarse search results and the remote resultant web indices.

7. The method of claim 6, further comprising:

transmitting the at least one best candidate to the remote voice search processing unit; and

receiving information specific to the at least one best candidate from the remote voice search processing unit.

8. The method of claim 1, wherein the fine search results contain a plurality of best candidates from the coarse search results and the remote resultant web indices, and the plurality of best candidates are displayed to the user.

9. The method of claim 8, further comprising:

receiving a selection of at least one of the plurality of best candidates from the user;

transmitting the user'"'"'s selection to the remote voice search processing unit; and

receiving information specific to the user'"'"'s selection from the remote voice search processing unit.

10. A mobile communication device, comprising:

a voice search engine that receives a search query from a user of the mobile communication device, generates a lattice of coarse linguistic representations from speech parts in the search query, extracts query features from the generated lattice of coarse linguistic representations, generates coarse search feature vectors based on the extracted query features, and performs a coarse search using the generated coarse search feature vectors;

a transceiver that transmits the generated coarse search feature vectors to a remote voice search processing unit and receives remote resultant web indices from the remote voice search processing unit based on the generated coarse search feature vectors, wherein the voice search engine generates a lattice of fine linguistic representations from speech parts in the search query based on the coarse search results and the remote resultant web indices, generates fine search feature vectors from the lattice of fine linguistic representations, and performs a fine search using the coarse search results, the remote resultant web indices and the generated fine search feature vectors; and

a display that displays the fine search results to the user.

11. The mobile communication device of claim 10, wherein the coarse linguistic representations and the fine linguistic representations are at least one of words, phrases, morphemes, syllables, and phonemes, of at least one language.

12. The mobile communication device of claim 10, further comprising:

a metadata database that stores multimedia content with linguistic metadata, wherein the coarse search feature vectors and the fine search feature vectors are generated using the linguistic metadata from the metadata database, the multimedia content being least one of audio files, audio recordings, emails, voice mails, recorded conversations, notes, messages, text messages, photos, video files, movie files, and television recordings.

13. The mobile communication device of claim 10, wherein the remote resultant web indices result from at least one search conducted by at least one web search engine.

14. The mobile communication device of claim 10, wherein the coarse search results and the remote resultant web indices contain a plurality of candidate selections based on the coarse search feature vectors.

15. The mobile communication device of claim 10, wherein the fine search results contain at least one best candidate from a combination of the coarse search results and the remote resultant web indices.

16. The mobile communication device of claim 15, wherein the transceiver transmits the at least one best candidate to the remote voice search processing unit and receives information specific to the at least one best candidate from the remote voice search processing unit.

17. The mobile communication device of claim 10, wherein the fine search results contain a plurality of best candidates out of the coarse search results and the web indices, and the display displays the plurality of best candidates to the user.

18. The mobile communication device of claim 17, wherein the voice search engine receives a selection of at least one of the plurality of best candidates from the user, and the transceiver transmits the user'"'"'s selection to the remote voice search processing unit and receives information specific to the best candidate from the remote voice search processing unit.

19. The mobile communication device of claim 10, wherein the mobile communication device is one of a mobile telephone, cellular telephone, a wireless radio, a portable computer, a laptop, an MP3 player, satellite radio, and a satellite television.

20. A method of performing an Internet (web) search in response to a voice search query from a user of a mobile communication device, comprising:

receiving generated coarse search feature vectors from a mobile communication device;

performing a web search based on the received coarse search feature vectors using at least one web search engine;

receiving indexed web search results from the at least one web search engine; and

transmitting the indexed web search results and their linguistic representations to the mobile communication device.