Methods and systems for obtaining language models for transcribing communications

1. A method for transcribing a spoken communication, comprising:

receiving a spoken first communication from a first sender to a first recipient;

obtaining information relating to a second communication from a second sender to a second recipient, wherein the second communication was made before the spoken first communication and is different from the spoken first communication;

using the obtained information to create a final language model by:

obtaining a general language model;

using the obtained information to create a specific language model; and

creating the final language model using the general language model and the specific language model; and

using the language model to transcribe the spoken first communication.

2. The method of claim 1 , wherein the first communication is a voicemail message left by the first sender for the first recipient.

3. The method of claim 1 , wherein the first sender and the second sender are the same person.

4. The method of claim 1 , wherein the second communication is selected from a plurality of communications based on a proximity of the second communication to the first communication in a communication graph.

5. The method of claim 1 , wherein the second communication is selected from a plurality of communications based on a time of the second communication and a time of the first communication.

6. The method of claim 1 , further comprising using the obtained information to perform language identification.

7. A non-transitory computer-readable medium containing a program which, when executed, performs a method, the method comprising:

receiving a spoken first communication from a first sender to a first recipient;

obtaining information relating to a second communication from a second sender to a second recipient, wherein the second communication was made before the spoken first communication and is different from the spoken first communication;

using the obtained information to create a final language model by:

obtaining a general language model;

using the obtained information to create a specific language model; and

creating the final language model using the general model and the specific language model; and

using the language model to transcribe the spoken first communication.

8. The computer-readable medium of claim 7 , wherein the second communication is selected from a plurality of communications based on a relationship between the first sender and the second sender.

9. The computer-readable medium of claim 8 , wherein the relationship is one of (i) that the first sender and the second sender are coworkers, (ii) that the first sender and the second sender are relatives; and (iii) that the first sender and the second sender are friends.

10. The computer-readable medium of claim 7 , wherein the obtained information comprises information about the second sender or the second recipient.

11. The computer-readable medium of claim 7 , wherein the obtained information comprises a location of the second sender.

12. A system for transcribing a communication, comprising:

a device, including:

a memory containing a program for transcribing speech; and

a processor which, when executing the program, performs an operation comprising:

receiving a spoken first communication from a first sender to a first recipient;

obtaining information relating to a second communication from a second sender to a second recipient, wherein the second communication was made before the spoken first communication and is different from the first spoken communication;

using the obtained information to create a final language model by:

obtaining a general language model;

using the obtained information to create a specific language model; and

creating the final language model using the general language model and the specific language model; and

using the language model to transcribe the spoken first communication.

13. The system of claim 12 , wherein the obtained information includes the time and date of the second communication.

14. The system of claim 12 , wherein the obtained information includes information about weather.

15. The system of claim 12 , wherein the obtained information includes information about a news event.

16. The system of claim 12 , wherein the obtained information includes information about the second sender and the time and date of the second communication.

17. The system of claim 12 , wherein obtaining information relating to the second communication includes retrieving information, from the Internet, relating to the second sender or to the second recipient.