Investigating Confidence Estimation Measures for Speaker Diarization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914848199147520 |
|---|---|
| author | Chowdhury, Anurag Misra, Abhinav Fuhs, Mark C. Woszczyna, Monika |
| author_facet | Chowdhury, Anurag Misra, Abhinav Fuhs, Mark C. Woszczyna, Monika |
| contents | Speaker diarization systems segment a conversation recording based on the speakers' identity. Such systems can misclassify the speaker of a portion of audio due to a variety of factors, such as speech pattern variation, background noise, and overlapping speech. These errors propagate to, and can adversely affect, downstream systems that rely on the speaker's identity, such as speaker-adapted speech recognition. One of the ways to mitigate these errors is to provide segment-level diarization confidence scores to downstream systems. In this work, we investigate multiple methods for generating diarization confidence scores, including those derived from the original diarization system and those derived from an external model. Our experiments across multiple datasets and diarization systems demonstrate that the most competitive confidence score methods can isolate ~30% of the diarization errors within segments with the lowest ~10% of confidence scores. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_17124 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Investigating Confidence Estimation Measures for Speaker Diarization Chowdhury, Anurag Misra, Abhinav Fuhs, Mark C. Woszczyna, Monika Sound Machine Learning Audio and Speech Processing Speaker diarization systems segment a conversation recording based on the speakers' identity. Such systems can misclassify the speaker of a portion of audio due to a variety of factors, such as speech pattern variation, background noise, and overlapping speech. These errors propagate to, and can adversely affect, downstream systems that rely on the speaker's identity, such as speaker-adapted speech recognition. One of the ways to mitigate these errors is to provide segment-level diarization confidence scores to downstream systems. In this work, we investigate multiple methods for generating diarization confidence scores, including those derived from the original diarization system and those derived from an external model. Our experiments across multiple datasets and diarization systems demonstrate that the most competitive confidence score methods can isolate ~30% of the diarization errors within segments with the lowest ~10% of confidence scores. |
| title | Investigating Confidence Estimation Measures for Speaker Diarization |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2406.17124 |