Speech-recognition-assisted selective suppression of silent and filled speech pauses during playback of an audio recording

• US 6,161,087 A
• Filed: 10/05/1998
• Issued: 12/12/2000
• Est. Priority Date: 10/05/1998
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Lernout & Hauspie Speech Products NV (Intel Corporation)
• Inventors: Bachenko, Joan, Wightman, Colin W.
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Smits, Talivaldis Ivars
• Application Number:
  US09/166,364
• Time in Patent Office: 799 Days
• Field of Search: 704/215, 704/233, 704/251, 704/278
• US Class Current: 704/215
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2025/783 based on threshold decision

Abstract

A method for playback of speech in an audio recording. The method comprises performing full word-level recognition of the speech including recognition of silent pauses and filled pauses, suppressing playback of the filled pauses and silent pauses, alerting a listener of the audio recording to locations of suppressed filled pauses and silent pauses during playback of the audio recording, and accepting a user command to disable suppression of any filled pause or silent pause during playback of the audio recording.

Claims

1. A method for playback of speech in an audio recording, the method comprising:

(a) performing full word-level recognition of the speech including recognition of one or more silent pauses or filled pauses;

(b) suppressing playback of one or more filled pauses or silent pauses;

(c) alerting a listener of the audio recording to locations of suppressed filled pauses or silent pauses during playback of the audio recording; and

(d) accepting a user command to disable suppression of one or more filled pauses or silent pauses during playback of the audio recording.

2. The method of claim 1 wherein the act of alerting comprises using a visual signal.

3. The method of claim 1 wherein the act of alerting comprises using an audible signal.

4. The method of claim 1 wherein the act of performing recognition further comprises associating a start and a stop time with one or more silent pause or filled pause in the speech.

5. The method of claim 4 wherein the act of performing recognition further comprises identifying pauses that are of greater length than a minimum pause suppression value.

6. The method of claim 5 wherein the act of performing recognition further comprises adjusting both the start time and the stop time of each identified pause by a guard period to decrease the length of the identified pause, and wherein the act of suppressing comprises suppressing playback of only each identified pause.

7. A method for controlling playback of filled pauses during playback of speech in audio recordings, the method comprising:

(a) performing full word-level recognition of the speech including recognition of filled pauses, wherein the act of performing recognition includes recognizing filled pauses as actual filled pauses only if a pattern of occurrence of the recognized filled pauses matches with a pattern of occurrence of filled pauses in speech; and

(b) suppressing playback of the actual filled pauses.

8. The method of claim 7 further comprising alerting a listener of the audio recording to locations of suppressed actual filled pauses during playback of the audio recording.

9. The method of claim 8 wherein the act of alerting comprises using a visual signal.

10. The method of claim 8 wherein the act of alerting comprises using an audible signal.

11. The method of claim 8 further comprising allowing a user to disable suppression of one or more actual filled pauses during playback of the audio recording.

12. The method of claim 7 wherein the act of performing recognition further comprises identifying pauses that are of greater length than a minimum pause period and adjusting both a start time and a stop time of each identified pause by a guard period to decrease the length of the identified pause prior to performing act (b).

13. The method of claim 12 wherein the act of suppressing comprises suppressing playback of only each identified pause.

14. An apparatus for playback of speech in an audio recording, the apparatus comprising:

(a) a speech recognizer to perform full word-level recognition including recognition of silent pauses and filled pauses;

(b) software containing instructions to suppress playback of silent pauses and filled pauses; and

(c) a user interface to alert a listener of the audio recording to locations of suppressed filled pauses and silent pauses during playback of the audio recording and to allow a user to disable suppression of one ore more filled pauses or silent pauses during playback of the audio recording.

15. The apparatus of claim 14 wherein the user interface comprises an audible signal.

16. The apparatus of claim 14 wherein the user interface comprises a visual signal.

17. The apparatus of claim 14 wherein the software further contains instructions to treat filled pauses recognized by the speech recognizer as actual filled pauses only if a pattern of occurrence of the recognized filled pauses matches with a pattern of occurrence of filled pauses in speech, and wherein the instructions suppress playback of only actual filled pauses.

18. The apparatus of claim 14 wherein the software further contains instructions to identify pauses that are of greater length than a minimum pause suppression value, and wherein the instructions suppress playback of only the so-identified pauses.

19. The apparatus of claim 18 wherein the software further contains instructions to adjust at least one of a start time and a stop time of each identified pause by a guard period to decrease the length of each identified pause.

20. The apparatus of claim 14 further comprising a suppressing processor to identify by specified criteria and to mark pauses for suppression during playback.