System and method for synchronizing sound and manually transcribed text

1. A method for synchronizing recorded sound data (SD) and text data (TD), said text data (TD) being obtained by manual transcription of said recorded sound data (SD) during playback of the latter, the method comprising:

repeatedly and substantially continuously during said playback performing the steps of:

querying, via a computer executing stored program instructions, said recorded, played back sound data (SD) and said text data (TD) to obtain at a current time position (TPi) a corresponding currently played sound datum (SDi) and a currently transcribed text datum (TDj),

generating, via a computer executing stored program instructions, for the TDj, a corrected time position (TPi′) by applying a time correction value (L) in accordance with a transcription delay to the current time position (TPi),

generating via a computer executing stored program instructions at least one association datum (<TPi′>) indicative of a synchronization association between said corrected time position (TPi′) and said currently transcribed text datum (TDj), and

in response to identifying a pause in said sound data (SD) that corresponds to punctuation in said text data (TD), generating an additional association datum (<TPX′>).

2. The method according to claim 1 , wherein said association data (<TPi′>, <TPX′>) are stored together with said text data (TD) in a common synchronized text data file (TDF).

3. The method according to claim 1 , wherein said association data (<TPi′>, <TPX′>) are stored separately from said text data (TD) in a synchronization file.

4. A method for synchronized playback of sound data (SD) and corresponding text data (TD), comprising the steps of repeatedly:

playing back a respective sound datum (SDi) at a given point of time (t), and

showing a text datum (TDj) associated with that sound datum (SDi) at substantially said same point of time (t),

wherein the associated text datum (TDj) is obtained by the method according to claim 1 .

5. The method according to claim 4 , wherein the text datum (TDj) corresponding particular pauses in said sound data (SD) corresponding to punctuation in said text data (TD), is shown at essentially a point of time (t) at which the sound data (SD) meets a predetermined criterion indicative of said pauses.

6. The method of claim 1 , wherein the steps of the method are performed repeatedly and substantially continuously at intervals on the order of about 0.1-1 second.

7. The method of claim 1 , wherein the additional association datum (<TPX′>) represents a time position (TPX′) associated with the identified pause.

8. A system for synchronizing sound data (SD) and text data (TD), comprising:

sound input means for providing said sound data (SD), in particular in the form of human speech;

storage means for storing said sound data (SD);

playback means for playing back said sound data (SD) to a user;

text input means for manual transcription of said sound data (SD) by providing said text data (TD); and

data processing means for creating association data (<TPi′>, <TPX′>) indicative of a synchronization association between said sound data (SD) and said text data (TD), said data processing means comprising:

query means for repeatedly and substantially continuously querying said sound data (SD) and said text data (TD) to obtain at a current time position (TPi) a corresponding currently played sound datum (SDi) and a currently entered text datum (TDj),

correcting means for, repeatedly and substantially continuously generating, for the TDj, a corrected time position (TPi′) by applying a time correction value (L) in accordance with a transcription delay to the current time position (TPi), and

data generating means for, repeatedly and substantially continuously generating an association datum (<TPi′>) indicative of a synchronization association between said corrected time position (TPi′) and said currently entered text datum (TDj),

wherein said data processing means are adapted for identifying pauses in said sound data (SD) corresponding to punctuation in said text data (TD), and for generating additional association datum (<TPX′>).

9. The system according to claim 8 , wherein said association data (<TPi′>, <TPX′>) are stored together with said text data (TD) in a common synchronized text data file (TDF).

10. The system according to claim 8 , wherein said association data (<TPi′>, <TPX>) are stored separately from said text data (TD) in a synchronization file.

11. The system according to claim 8 , further for synchronized playback of sound data (SD) and corresponding text data (TD), comprising:

playback means for playing back a respective sound datum (SDi) at a given point of time (t), and

showing means for showing a text datum (TDj) associated to that sound datum (SDi) at substantially said same point of time (t).

12. The system according to claim 11 , wherein said data processing means ( 6 ) are adapted for identifying pauses in said sound data (SD) corresponding to punctuation in said text data (TD), and for commanding said showing means to show a respective text datum (TDj) corresponding to a said pause at essentially a point of time (t) at which said sound data (SD) meets a predetermined criterion indicative of said pause.

13. A method comprising using the system according to claim 12 for the transcription of sound to text, in particular for medical transcription.

14. A method comprising using the system according to claim 12 as part of a correction stage in a document creation workflow, the latter comprising at least the stages of:

dictation,

speech recognition, and

correction.

15. The system of claim 8 , wherein:

the query means repeatedly and substantially continuously queries said sound data (SD) and said text data (TD) at intervals on the order of about 0.1-1 second;

the correcting means repeatedly and substantially continuously generate the corrected time position (TPi′) at intervals on the order of about 0.1-1 second; and

the data generating means repeatedly and substantially continuously generate the association datum (<TPi′>) at intervals on the order of about 0.1-1 second.

16. The system of claim 8 , wherein the additional association datum (<TPX′>) represents a time position (TPX') associated with the identified pause.