Personalized user specific grammars

• US 7,693,267 B2
• Filed: 12/30/2005
• Issued: 04/06/2010
• Est. Priority Date: 12/30/2005
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Seide, Frank Torsten Bernd, Howell, David Andrew, Fong, David T., Sundararaman, Sridhar
• Primary Examiner(s): Tsang; Fan
• Assistant Examiner(s): Teshale; Akelaw A
• Application Number:
  US11/324,208
• Publication Number:
  US 20070153989A1
• Time in Patent Office: 1,558 Days
• Field of Search: 379/88.14, 704/209, 455/414.4
• US Class Current: 379/88.14
• CPC Class Codes: H04M 2201/405 involving speaker-dependent recognition
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 2203/253 where a visual mode is used instead of a voice mode
• CPC Class Codes: H04M 2203/4509 Unified messaging with single point of access to voicemail and other mail or messaging systems
• CPC Class Codes: H04M 3/5307 for recording messages comprising any combination of audio and non-audio components

Abstract

Improved systems and methods are provided for transcribing audio files of voice mails sent over a unified messaging system. Customized grammars specific to a voice mail recipient are created and utilized to transcribe a received voice mail by comparing the audio file to commonly utilized words, names, acronyms, and phrases used by the recipient. Key elements are identified from the resulting text transcription to aid the recipient in processing received voice mails based on the significant content contained in the voice mail.

Claims

1. A computer-implemented method for improved transcription to a text file of an audio file of a voice mail message to a recipient in a unified messaging system comprising a grammar and a key elements list, the method comprising:

comparing each voice mail utterance on the audio file to the grammar;

determining one or more voice mail utterances as recognizable based on a threshold relating to a probability accuracy of said comparing;

transcribing each of said recognizable voice mail utterances into a text corresponding to each said recognizable voice mail utterance, wherein the plurality of texts comprise a text file;

determining one or more voice mail utterances as unrecognizable based on a threshold relating to a probability accuracy of said comparing;

representing each of said unrecognizable voice mail utterances as a symbol or string added to the text file; and

including with the text file an identification of key elements in the texts in the text file that are contained in the key elements list.

2. The computer-implemented method of claim 1, wherein the grammar is a personal grammar and wherein the personal grammar is specific to the recipient based on at least one of electronic-mail (email) inbox, voice mail inbox, telephone call logs, email directory, telephone directory, contacts list, appointments calendar, organizational charts of the recipient and documents specific to an individual.

3. The computer-implemented method of claim 1, wherein the grammar is a personal grammar and the unified messaging system further comprising a second grammar, wherein the combination of the personal grammar and the second grammar are used to transcribe the audio file to a text file.

4. The computer-implemented method of claim 3, wherein the second grammar is selected from the group consisting of voice mail sender personal grammars, subgroup grammars, hierarchical grammars, common unified messaging system grammars, common security clearance grammars, organizational grammars, and occupational grammars.

5. The computer-implemented method of claim 1, wherein a message is sent to the recipient comprising the identified key elements, the transcribed text, and the voice mail audio file.

6. The computer-implemented method of claim 5, wherein an identified key element is provided in a subject line or header line of the message.

7. The computer-implemented method of claim 5, wherein the identified key elements are grouped into categories.

8. The computer-implemented method of claim 7, wherein the categories are selected from the group consisting of subject matter, action item, phone numbers, and scheduled times.

9. The computer-implemented method of claim 5, wherein each key element in the message is hyperlinked to an audio file location within the voice mail audio file in the message where the key element is uttered.

10. The computer-implemented method of claim 9, further comprising:

activating the key element hyperlink to playback the utterance of the key element in the voice mail audio file.

11. The method of claim 1, wherein one or more computer-readable media have computer-executable instructions for performing the computer-executable method of claim 1.

12. A system for transcribing an audio file voice mail that is sent to a recipient connected to a unified messaging system, comprising:

a grammar containing data from the unified messaging system mailbox of the recipient;

a key elements list containing data that is to be located in a voice mail;

a server being configured to execute computer-executable instructions for;

executing a speech-to-text module that compares each voice mail utterance on the audio file to the grammar;

creating a text corresponding to each recognizable voice mail utterance, wherein the plurality of texts comprise a text transcription;

determining one or more utterances in the audio file as unrecognizable based on a threshold relating to a probability accuracy of said comparing, said determining further comprising representing each of said unrecognizable utterances as a symbol or string in a corresponding position within the text transcription;

identifying key elements contained in the text transcription that are contained in the key elements list, said identified key elements corresponding to only the recognizable voice mail utterances from which a text has been created; and

sending the recipient an email containing the identified key elements, the text transcription, and the audio file.

13. The system of claim 12, wherein the grammar is a personal grammar and wherein the speech-to-text module transcribes the audio file based on the personal grammar and at least one additional grammar.

14. The system of claim 13, wherein the additional grammar is selected from the group consisting of voice mail sender personal grammars, subgroup grammars, hierarchical grammars, common unified messaging system grammars, common security clearance grammars, organizational grammars, and occupational grammars.

15. The system of claim 12, wherein the key elements identified in the email are hyperlinked to the audio file location of the key elements.

16. The system of claim 15, further comprising:

a computing device connected to a unified messaging system and having access to the recipient'"'"'s mailbox, the computing device being configured to execute a media player in response to activation of a key element hyperlink.

17. The system of claim 12, wherein the identified key elements are organized in categories in the email.

18. The system of claim 17, wherein the categories are selected from the group consisting of subject matter, action item, phone numbers, and scheduled times, and wherein the key elements contained in the email are color coded according to their categories.