Transcription system

• US 9,774,747 B2
• Filed: 04/29/2011
• Issued: 09/26/2017
• 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
• Primary Examiner(s): Sharma, Neeraj
• Application Number:
  US13/097,875
• Publication Number:
  US 20120278071A1
• Time in Patent Office: 2,342 Days
• Field of Search: 704235, 704234, 704 9, 704260, 704275, 704211, 704236, 704 1, 707750, 715704, 715728, 379 8801, 369 2501, 725 1
• 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 computer implemented playback unit for playing an audio recording to a user at a controlled playback rate;

a computer implemented keyboard unit, including an input device for accepting transcribed word input from the user;

a word spotter configured to receive transcribed words accepted at the input device of the keyboard unit, locate the transcribed words in the audio recording, and estimate corresponding audio times of the transcribed words in the audio recording; and

a computer implemented playback control unit configured to

(a) repeatedly determine a keyboard delay, the keyboard delay representing a difference between an audio time of the word and a current position of the playback unit, and

(b) control a playback rate of the audio recording according to the determined keyboard delay and a predetermined target delay, including controlling the playback rate to match the keyboard delay to the predetermined target delay, the controlling including slowing down the playback rate when the keyboard delay is greater than the target delay and increasing the playback rate when the keyboard delay is less than the target delay.

11. The method of claim 1 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.

12. The method of claim 11 wherein providing the predicted transcription includes predicting selected words from the transcribed words, and determining whether the selected words are present in the following audio.

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

14. The method of claim 11 wherein providing the predicted transcription include providing multiple alternative predictions for selection by the user.

15. The method of claim 12 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.

2. 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 including accepting transcribed words;

determining a correspondence of the accepted transcription and the audio recording including locating, by a word spotter, the transcribed words in the audio recording;

determining a keyboard delay for an accepted transcribed word, the keyboard delay representing a difference between a playback time of the word and a current playback position of the audio recording, said playback time being determined according to the locating of the transcribed words in the audio recording determined by the word spotter; and

controlling a playback rate according to the keyboard delay and a predetermined target delay, including applying a feedback approach to control playback of the audio recording to match the keyboard delay to the predetermined target delay, the controlling including both increasing and decreasing the playback rate according to the keyboard delay and the target delay.

3. The method of claim 2 wherein locating the words includes applying a wordspotting approach.

4. The method of claim 3 wherein applying the wordspotting approach includes using a stored phonetic processing of the audio recording determined prior to the playing of the recording.

5. The method of claim 2 wherein controlling the playback rate includes at least one of pausing the playback and rewinding the playback according to the determined delay.

6. The method of claim 2 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.

7. The method of claim 6 wherein providing the prediction includes predicting words from the transcription, and determining whether the words are present in the following audio.

8. The method of claim 7 wherein determining whether the words are present includes applying a wordspotting approach.

9. The method of claim 6 wherein providing the prediction include providing multiple alternative predictions for selection by the user.

10. The method of claim 7 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.