Transcription analysis platform
Various embodiments of the present disclosure evaluate transcription accuracy. In some implementations, the system normalizes a first transcription of an audio file and a baseline transcription of the audio file. The baseline transcription can be used as an accurate transcription of the audio file. The system can further determine an error rate of the first transcription by aligning each portion of the first transcription with the portion of the baseline transcription, and assigning a label to each portion based on a comparison of the portion of the first transcription with the portion of the baseline transcription.
1. A method comprising:
receiving, at a graphical user interface (GUI), a transcription of an audio file and a baseline transcription of the audio file;
displaying, by the GUI, one or more portions of the transcription aligned with one or more corresponding portions of the baseline transcription;
displaying, by the GUI, a differentiator label for each portion of the transcription based on a comparison of a portion of the transcription with a corresponding portion of the baseline transcription;
displaying, by the GUI, an error rate of the transcription based on determinations of whether each portion, as aligned between the transcription and the baseline transcription, has a differentiator label corresponding to correct, inserted, substituted in, or deleted respective differentiator labels assigned to each portion of the transcription; and
displaying, by the GUI, a heat map that is based on A) the error rate and B) determined word lengths of the transcription and/or the baseline transcription.
2. The method of claim 1 , further comprising:
displaying an inserted section into the transcription at a location where a deleted word was deleted from the transcription.
3. The method of claim 1 , further comprising:
displaying a normalization of the transcription and the baseline transcription by automatically changing words and/or numbers to a standardized spelling or appearance.
4. The method of claim 1 , wherein an x-axis on the heat map represents the error rate and a y-axis represents a length difference of the transcription compared to the baseline transcription.
5. The method of claim 1 , wherein an x-axis on the heat map represents a word accuracy rate and a y-axis represents a length of a call according to the baseline transcription.
6. The method of claim 1 , wherein the error rate is determined by dividing a number of correct portions by a sum of at least a number of portions with a label corresponding to inserted, a number of portions with a label corresponding to deleted, and a number of portions with a label corresponding to substituted.
7. The method of claim 1 , wherein the error rate is an error rate of incorrect phrases between the transcription and the baseline transcription.
8. A system comprising:
one or more processors; and
one or more memories storing instructions that, when executed by the one or more processors, cause the system to perform a process comprising:
receiving, at a graphical user interface (GUI), a transcription of an audio file and a baseline transcription of the audio file;
displaying, by the GUI, one or more portions of the transcription aligned with one or more corresponding portions of the baseline transcription;
displaying, by the GUI, a differentiator label for each portion of the transcription based on a comparison of a portion of the transcription with a corresponding portion of the baseline transcription;
displaying, by the GUI, an error rate of the transcription based on determinations of whether each portion, as aligned between the transcription and the baseline transcription, has a differentiator label corresponding to correct, inserted, substituted in, or deleted respective differentiator labels assigned to each portion of the transcription; and
displaying, by the GUI, a heat map that is based on A) the error rate and B) determined word lengths of the transcription and/or the baseline transcription.
9. The system according to claim 8 , wherein the process further comprises:
displaying an inserted section into the transcription at a location where a deleted word was deleted from the transcription.
10. The system according to claim 8 , wherein the process further comprises:
displaying a normalization of the transcription and the baseline transcription by automatically changing words and/or numbers to a standardized spelling or appearance.
11. The system according to claim 8 , wherein an x-axis on the heat map represents the error rate and a y-axis represents a length difference of the transcription compared to the baseline transcription.
12. The system according to claim 8 , wherein an x-axis on the heat map represents a word accuracy rate and a y-axis represents a length of a call according to the baseline transcription.
13. The system according to claim 8 , wherein the error rate is determined by dividing a number of correct portions by a sum of at least a number of portions with a label corresponding to inserted, a number of portions with a label corresponding to deleted, and a number of portions with a label corresponding to substituted.
14. The system according to claim 8 , wherein the error rate is an error rate of incorrect phrases between the transcription and the baseline transcription.
15. A non-transitory computer-readable medium storing instructions that, when executed by a computing system, cause the computing system to perform operations comprising:
receiving, at a graphical user interface (GUI), a transcription of an audio file and a baseline transcription of the audio file;
displaying, by the GUI, one or more portions of the transcription aligned with one or more corresponding portions of the baseline transcription;
displaying, by the GUI, a differentiator label for each portion of the transcription based on a comparison of a portion of the transcription with a corresponding portion of the baseline transcription;
displaying, by the GUI, an error rate of the transcription based on determinations of whether each portion, as aligned between the transcription and the baseline transcription, has a differentiator label corresponding to correct, inserted, substituted in, or deleted respective differentiator labels assigned to each portion of the transcription; and
displaying, by the GUI, a heat map that is based on A) the error rate and B) determined word lengths of the transcription and/or the baseline transcription.
16. The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:
displaying an inserted section into the transcription at a location where a deleted word was deleted from the transcription.
17. The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:
displaying a normalization of the transcription and the baseline transcription by automatically changing words and/or numbers to a standardized spelling or appearance.
18. The non-transitory computer-readable medium of claim 15 , wherein an x-axis on the heat map represents the error rate and a y-axis represents a length difference of the transcription compared to the baseline transcription.
19. The non-transitory computer-readable medium of claim 15 , wherein an x-axis on the heat map represents a word accuracy rate and a y-axis represents a length of a call according to the baseline transcription.
20. The non-transitory computer-readable medium of claim 15 , wherein the error rate is determined by dividing a number of correct portions by a sum of at least a number of portions with a label corresponding to inserted, a number of portions with a label corresponding to deleted, and a number of portions with a label corresponding to substituted.