Transcription service stopping automatic transcription
View Patent ↗A transcription system including a transcription device for the automatic transcription of dictated material and additionally employs transcribers who manually transcribe some of the dictated material. The transcription device contains an abortion unit for aborting processing of certain dictated material by the transcription device and retransmitting this dictated material to one of the transcribers for manual transcription if the quality of the dictated audio data (AI) or the quality of the automatically transcribed text data (ATTI) is so poor that the correction of recognition errors in the automatically transcribed text data (ATTI) requires more effort than if said dictated material is transcribed manually from the outset.
1. A transcription device for the automatic transcription of dictated material, comprising:
receiving means for receiving dictated audio data (AI);
speech recognition means for converting the received audio data (AT) into automatically transcribed text data (ATTI);
output means for outputting the automatically transcribed text data (ATTI); and
abortion means for aborting processing of the dictated material by the transcription device if automatic transcription by the speech recognition means would produce or has already produced poor quality automatically transcribed text data (ATTI) with a relatively large number of recognition errors.
2. A transcription device as claimed in claim 1 , in which the abortion means are designed to check the quality of the received audio data (AI) for downstream speech recognition by the speech recognition means and to abort processing of the dictated material by the transcription device if the quality of the received audio data (AI) does not allow successful automatic transcription.
3. A transcription device as claimed in claim 2 , in which the abortion means are designed to check the signal-to-noise ratio of an audio signal containing the received audio data (AI).
4. A transcription device as claimed in claim 2 , in which the abortion means are designed to check the level of the audio signal containing the received audio data (AI).
5. A transcription device as claimed in claim 2 , in which the abortion means are designed to check whether the audio signal containing the received audio data has non-linear distortions.
6. A transcription device as claimed in claim 1 , in which the abortion means are designed to check the quality of the automatically transcribed text data (ATTI) and to abort processing of the dictated material by the transcription device if the automatically transcribed text data (ATTI) contains a number exceeding an error threshold of recognition errors per N words of the automatically transcribed text data (ATTI).
7. A transcription device as claimed in claim 1 , in which retransmission means are provided which are designed to retransmit the received audio data (AI) after processing of the dictated material by the transcription device has been aborted, in order to allow manual transcription of the dictated material.
8. A transcription method for the automatic transcription of dictated material, in which the following steps are executed:
reception of dictated audio data (AI);
conversion of the received audio data (AI) into automatically transcribed text data (ATTI);
output of the automatically transcribed text data (ATTI);
abortion of processing of the dictated material if the automatically transcribed text data (ATTI) produced during automatic transcription would be or already is of poor quality with a relatively large number of recognition errors.
9. A transcription method as claimed in claim 8 , in which the following further steps are executed:
checking of the quality of the received audio data (AI) for downstream speech recognition in which the signal-to-noise ratio of an audio signal containing the received audio data (AI) is checked;
abortion of processing of the dictated material if the quality of the received audio data (AI) does not allow successful automatic transcription.
10. A transcription method as claimed in claim 8 , in which the following further steps are executed:
checking of the quality of the received audio data (AI) for downstream speech recognition in which the level of the audio signal containing the received audio data (AI) is checked;
abortion of processing of the dictated material if the quality of the received audio data (AI) does not allow successful automatic transcription.
11. A transcription method as claimed in claim 8 , in which the following further steps are executed:
checking of the quality of the automatically transcribed text data (ATTI);
abortion of processing of the dictated material if the automatically transcribed text data contains a number exceeding an error threshold of recognition errors per N words of the automatically transcribed text data (ATTI).
12. A transcription method as claimed in claim 8 , in which the following further steps are executed:
retransmission of the received audio data (AI), in order to allow manual transcription of the dictated material if processing of the dictated material has been aborted.
13. A computer readable medium, which may be loaded directly into the internal memory of a digital computer and comprises software code sections, causing said digital computer to execute the steps of the transcription method as claimed in claim 8 are executed by the computer if the product is running on the computer.
14. A computer program product as claimed in claim 13 , in which the following further steps are executed:
checking of the quality of the received audio data (AI) for downstream speech recognition in which the signal-to-noise ratio of an audio signal containing the received audio data (AI) is checked;
abortion of processing of the dictated material if the quality of the received audio data (AI) does not allow successful automatic transcription.