Systems and methods for punctuating voicemail transcriptions

• US 8,719,004 B2
• Filed: 03/19/2010
• Issued: 05/06/2014
• Est. Priority Date: 03/19/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Ditech Networks, Inc. (Microsoft Corporation)
• Inventors: Siminoff, James
• Primary Examiner(s): Godbold, Douglas
• Assistant Examiner(s): ORTIZ SANCHEZ, MICHAEL
• Application Number:
  US12/727,736
• Publication Number:
  US 20100241429A1
• Time in Patent Office: 1,509 Days
• Field of Search: 704/6, 704/257, 704/270
• US Class Current: 704/6
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system, method and software product punctuates voicemail transcription text. A transcription text of the voicemail message is generated and the pauses between words of the transcribed text are determined. Ellipses are inserted into the transcription text at the position of “er” and “ahh” type words and pauses between words of the transcribed text.

Claims

1. A method for punctuating voicemail transcription text, comprising:

receiving, at a device, a voicemail message from telecommunication service provider;

transcribing, via the device, the voicemail into a transcription text and determining a beginning time and an ending time of each transcribed word within the voicemail;

identifying Y largest pauses between transcribed words of the transcription text, where Y is a limit on the number of ellipses to be inserted; and

inserting ellipses into the transcription text at the position of the Y largest pauses wherein the Y largest pauses is a number smaller than a total number of pauses of the transcription text.

2. The method of claim 1, wherein the pauses include sounds.

3. The method of claim 1, wherein the pauses include hesitations.

4. The method of claim 1, further comprising storing the transcription text at a storage device.

5. A software product comprising instructions, stored on non-transitory computer-readable media, wherein the instructions, when executed by a computer, perform steps for punctuating voicemail transcription text, comprising:

instructions for receiving, at a device, a voicemail message from telecommunication service provider;

instructions for transcribing, via the device, the voicemail into a transcription text and determining a beginning time and an ending time of each transcribed word within the voicemail;

instructions for identifying Y largest pauses within the transcription text, where Y is a limit on the number of ellipses to be inserted; and

instructions for inserting ellipses into the transcription text at the position of the Y largest pauses wherein the Y largest pauses is a number smaller than a total number of pauses of the transcription text.

6. The software product of claim 5, wherein the pauses include sounds.

7. The software product of claim 5, wherein the pauses include hesitations.

8. The software product of claim 5, further comprising storing the transcription text at a storage device.

9. A system for punctuating voicemail transcription text, comprising:

means for receiving a voicemail at a device;

means for transcribing, via the device, audio of the voicemail into transcription text and determining the beginning and finishing times of each transcribed word; and

means for identifying the Y largest pauses between transcribed words of the transcription text, where Y is a limit on the number of ellipses to be inserted; and

means for inserting ellipses into the transcription text at the position of the Y largest pauses wherein the Y largest pauses is a number smaller than a total number of pauses of the transcription text.

10. The system of claim 9, wherein the pauses include sounds.

11. The system of claim 9, wherein the pauses include hesitations.

12. The system of claim 9, further comprising storing the transcription text at a storage device.

13. A method for punctuating voicemail transcription text, comprising:

receiving, at a device, a voicemail message from telecommunication service provider;

transcribing, via the device, the voicemail into a transcription text and determining a beginning time and an ending time of each transcribed word within the voicemail;

identifying Y pauses within the transcription text, where Y is a limit on the number of ellipses to be inserted; and

inserting punctuation between a first word and a second word of the transcription text at the position of the Y pauses in order to indicate a hesitation, wherein the Y largest pauses is a number smaller than a total number of pauses of the transcription text.