TRANSCRIPTION SYSTEM

• US 20120278071A1
• Filed: 04/29/2011
• Published: 11/01/2012
• Est. Priority Date: 04/29/2011
• Status: Active Grant

• Current Assignee: Nexidia Incorporated (Nice Ltd)
• Original Assignee: Nexidia Incorporated (Nice Ltd)
• Inventors: Garland, Jacob B., Gavalda, Marsal
• Granted Patent: US 9,774,747 B2
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: H04M 11/10 with dictation recording and playback systems

Abstract

A transcription system automates the control of the playback of the audio to accommodate the user'"'"'s ability to transcribe the words spoken. In some examples, a delay between playback and typed input is estimated by processing the typed words using a wordspotting approach. The estimated delay is used as in input to an automated speed control, for example, to maintain a target or maximum delay between playback and typed input.

Claims

1. A transcription system comprising:

a controllable playback unit for playing an audio recording to a user;

a keyboard unit, including an input device for accepting word input from a user, and coupled to the audio player to accept a time reference from the player;

a word spotter configured to receive words from the keyboard unit and estimate corresponding audio times of the words in the audio recording; and

a playback control unit configured to control the playback unit according to the estimated audio times of the words accepted from the user.

2. The system of claim 1 wherein the playback control unit is configured to maintain a target delay between playback of words in the audio recording and the time the words are accepted for the user.

3. A method for controlling a transcription system comprising:

playing an audio recording to a user at a controlled rate;

accepting transcription of the audio recording from the user;

determining a correspondence of the accepted transcription and the audio recording; and

controlling a rate of playback according to the determined correspondence.

4. The method of claim 3 wherein accepting the transcription includes accepting transcribed words, and determining the correspondence includes locating the words in the audio recording.

5. The method of claim 4 wherein locating the words includes applying a wordspotting approach.

6. The method of claim 5 wherein applying the wordspotting approach includes using a stored phonetic processing of the audio recording determined prior to the playing of the recording.

7. The method of claim 4 wherein controlling the rate of playback includes determining a delay between a playback time of words and corresponding transcription times of the words.

8. The method of claim 7 wherein controlling the rate of playback includes applying a feedback approach based on the determined delay to approach a target delay.

9. The method of claim 7 wherein controlling the rate of playback includes at least one of pausing the playback and rewinding the playback according to the determined delay.

10. The method of claim 3 further comprising:

providing a predicted transcription to the user based on a speech recognition of following audio in an interval following a recording time corresponding to a transcription input from the user.

11. The method of claim 10 wherein providing the prediction includes predicting words from the transcription, and determining whether the words are present in the following audio.

12. The method of claim 11 wherein determining whether the words are present includes applying a wordspotting approach.

13. The method of claim 10 wherein providing the prediction include providing multiple alternative predictions for selection by the user.

14. The method of claim 11 further comprising providing a graphical representation of the predicted transcription to the user and updating the graphical representation based on the determination of whether the words are present in the following audio.

15. A transcription system comprising:

a storage for an audio recording;

a transcription entry unit for accepting word input from a user, and determining a time of occurrence of the words in the audio recording; and

a text prediction unit responsive to previously entered words and recorded audio configured to determined predicted words for transcription based on both the entered words and the recorded audio.

16. A transcription method comprising:

playing an audio recording to a user at a controlled rate;

accepting transcription of the audio recording from the user;

determining a correspondence of the accepted transcription and the audio recording; and

providing a predicted transcription to the user based on a speech recognition of the audio recording following times corresponding to the accepted transcription.

17. The method of claim 16 wherein accepting the transcription includes accepting transcribed words, and determined the correspondence includes locating the words in the audio recording.

18. The method of claim 17 wherein locating the words includes applying a wordspotting approach.

19. The method of claim 16 wherein providing the prediction includes predicting words from the transcription, and determining whether the words are present in the following audio.

20. The method of claim 19 wherein determining whether the words are present includes applying a wordspotting approach.

21. The method of claim 16 wherein providing the prediction include providing multiple alternative predictions for selection by the user.

22. A transcription method comprising:

accepting recorded media for transcription;

providing a graphical representation of the recorded media for transcription;

accepting a transcription of the recorded media from a user, the transcription including a plurality of transcription elements; and

updating the graphical representation of the recorded media based on the accepted transcription including updating future occurrences of the accepted transcription elements.

23. The method of claim 22 wherein the recorded media includes an audio recording and a video recording.

24. The method of claim 23 wherein the transcription elements are spoken words included in the audio recording.

25. The method of claim 23 wherein the transcription elements are the faces of people in the video recording.

26. The method of claim 22 wherein the recorded media includes an audio recording.

27. The method of claim 26, wherein the transcription elements are spoken words included in the audio recording.

28. The method of claim 22 wherein the graphical representation of the recorded media is a template.

29. The method of claim 28, wherein the graphical representation includes placeholders for words.

30. The method of claim 28 wherein the graphical representation includes placeholders for music.

31. The method of claim 28 wherein the graphical representation includes indicators of a change from one speaker to a different speaker.

32. The method of claim 28 wherein the template includes a plurality of elements and at least some of the plurality of elements includes editable metadata.

33. The method of claim 32 wherein the metadata includes a name of a speaker.

34. The method of claim 32 wherein the metadata includes a name of a piece of music.

35. The method of claim 22 wherein the graphical representation includes one or more indications of transcription completeness.

36. The method of claim 22 wherein the graphical representation includes one or more indications of transcription quality.

37. The method of claim 22 including automatically populating at least some of the graphical representation for recorded media with transcription elements.

38. The method of claim 37 wherein automatically populating at least some of the graphical representation includes applying a wordspotting approach.

39. The method of claim 22 wherein updating the graphical representation of the recorded media further includes detecting erroneous transcription elements included in the accepted transcription.