METHODS AND SYSTEMS FOR CORRECTING TRANSCRIBED AUDIO FILES

• US 20090276215A1
• Filed: 04/17/2007
• Published: 11/05/2009
• Est. Priority Date: 04/17/2006
• Status: Active Grant

• Current Assignee: III Holdings 1, LLC
• Original Assignee: Vovision, LLC (Microsoft Corporation)
• Inventors: Hager, Paul M.
• Granted Patent: US 8,407,052 B2
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G06F 16/31 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/0635 updating or merging of old and new templates; Mean values; Weighting

Abstract

Methods and systems for correcting transcribed text. One method includes receiving audio data from one or more audio data sources and transcribing the audio data based on a voice model to generate text data. The method also includes making the text data available to a plurality of users over at least one computer network and receiving corrected text data over the at least one computer network from the plurality of users. In addition, the method can include modifying the voice model based on the corrected text data.

Claims

1. A method of correcting transcribed text utilizing a computer-processing system, the computer-processing system having a browser-based user interface, the method comprising:

receiving audio data from one or more audio data sources;

transcribing the audio data based on a voice model to generate text data;

storing the text data;

making the text data available to a plurality of users over at least one computer network through the browser-based user interface;

receiving corrected text data over the at least one computer network from at least one of the plurality of users through the browser-based user interface; and

modifying the voice model based on the corrected text data.

2. The method of claim 1, wherein receiving audio data from one or more audio data sources includes receiving audio data from a VoIP voicemail server.

3. The method of claim 1, wherein receiving audio data from one or more audio data sources includes receiving audio data from a client computer over at least one computer network.

4. The method of claim 1, wherein receiving audio data from one or more audio data sources includes receiving audio data in an e-mail message.

5. The method of claim 1, wherein receiving audio data from one or more audio data sources includes requesting the audio data from the one or more audio data sources.

6. The method of claim 5, further comprising prioritizing the audio data.

7. The method of claim 1, wherein transcribing the audio data based on a voice model to generate text data includes transcribing the audio data based on a voice independent model to generate text data.

8. The method of claim 1, further comprising sending a correction notification to at least one of the plurality of users the correction notification including a link to access the text data through the browser-based interface.

9. The method of claim 8, wherein sending a correction notification to at least one of the plurality of users includes sending an e-mail correction notification to at least one of the plurality of users.

10. The method of claim 1, further comprising indexing the text data to create a keyword index the keyword index including a position of each of a plurality of keywords included in the text data.

11. The method of claim 1, further comprising sending a message notification to a user, the message notification including a link for accessing the text data through the browser based interface.

12. The method of claim 11, wherein sending a message notification to a user includes sending an e-mail message notification to the user.

13. The method of claim 1, further comprising delivering the corrected text data to at least one destination.

14. The method of claim 13, further comprising receiving the at least one destination from a user.

26. The method of claim 1 further comprising automatically identifying a plurality of speakers in the audio data based on changes in voice patterns in the audio data.

27. The method of claim 26, further comprising providing identifiers for each of the plurality of speakers in the text data.

28. The method of claim 1, further comprising automatically selecting a template for formatting the text data.

29. The method of claim 28, wherein automatically selecting a template for formatting the text data includes comparing the text data with one or more templates and automatically formatting the text data based on at least one template having a structure corresponding to a structure of the text data.

15. A system for correcting transcribed text, the system comprising:

a transcription server receiving audio data from one or more audio data sources;

at least one translation server to transcribe the audio data based on a voice model to generate text data,

a browser-based correction interface accessible by a plurality of users over at least one computer network and providing access to the text data and receiving corrected text data from at least one of the plurality of users; and

at least one training server receiving the corrected text data and modifying the voice model based on the corrected text data.

16. The system of claim 15, wherein the one or more audio data sources includes a VoIP voicemail server.

17. The system of claim 15, wherein the voice model includes a voice independent model.

18. The system of claim 15, wherein the transcription server sends a correction notification to at least one of the plurality of users including a link for accessing the text data through the browser-based correction interface.

19. The system of claim 18, wherein the correction notification includes an e-mail correction notification.

20. The system of claim 15, wherein the browser-based correction interface provides access to the audio data.

21. A method of performing audio data transcription utilizing a computer-processing system the computer-processing system having a browser-based interface, the method comprising:

obtaining audio data from at least one audio data source;

transcribing the audio data based on a voice-independent model to generate text data; and

sending an e-mail notification the text data to an owner of the audio data as an e-mail message the e-mail notification including a link to access the text data through the browser-based user interface.

22. The method of claim 21, further comprising sending a correction notification to at least one user after transcribing the audio data, the correction notification including a link to access the text data through the browser-based interface.

23. The method of claim 22, further comprising receiving corrected text data from the at least one user over at least one computer network through the browser-based interface.

24. The method of claim 23, further comprising modifying the voice independent model based on the corrected text data.

25. The method of claim 24, wherein the link allows the user to access the corrected text data through the browser-based interface.

30. A method of generating transcribed text data utilizing a computer-processing system, the computer-processing system having a streaming translation server, the method comprising:

receiving audio data from one or more audio data sources;

routing the audio data to the streaming translation server;

transcribing the audio data based on an independent voice model to generate text data in substantially real-time; and

providing the text data to one or more devices over at least one computer network.

31. The method of claim 30, wherein providing the text data to one or more devices includes providing the text data through a browser-based interface.

32. A method of training a voice independent model utilizing a computer-processing system, the computer-processing system having a voice independent model, a first transcription server, and a second transcription server, the method comprising:

transcribing audio data based on the voice independent model with the first transcription server to generate text data;

making the text data available to a user;

receiving corrected text data from the user;

creating an updated voice independent model based on the corrected text data and the voice independent model; and

sharing the updated voice independent model with the second transcription server.

33. A method of correcting transcribed text utilizing a computer-processing system, the computer-processing system having a browser-based interface, the method comprising:

receiving audio data from an audio data source;

transcribing a segment of the audio data to generate a corresponding text data segment;

making the text data segment available to a user through the browser-based interface;

receiving a corrected text data segment from the user through the browser-based interface; and

updating a voice independent model based on the corrected text data segment.