search

pyannote / pyannote-audio

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding
http://pyannote.github.io
MIT License
6.38k stars 784 forks source link

How to map the transcribed text with their respective speakers in speaker diarization? #1718

Closed ThiruRJST closed 5 months ago

ThiruRJST commented 6 months ago

Tested versions

Reproducible in 3.1

System information

Ubuntu-20.04

Issue description

I installed PyAnnote-3.1 and used speaker-diarization-3.1 for my project purpose. When I ran the pipeline, it gave me a rttm file as an output but the speakers were not mapped with their respective texts transcribed from it. Is there any workaround to achieve this?

Minimal reproduction example (MRE)

https://colab.research.google.com/github/pyannote/pyannote-audio/blob/develop/tutorials/MRE_template.ipynb

hbredin commented 6 months ago

pyannote does not do speech-to-text (what was said?). It only does speaker diarization (who spoke when?).

ThiruRJST commented 5 months ago

@hbredin Thank you for the confirmation