From 6aa947bc2fafdf6898d7f1a97f48c69a39a9d58a Mon Sep 17 00:00:00 2001 From: Subliminal Guy <subliminal_kid@posteo.de> Date: Sun, 1 Jun 2025 18:32:39 +0200 Subject: [PATCH] Add diarization info to README --- README.md | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/README.md b/README.md index 47fcc29..f57fcd4 100644 --- a/README.md +++ b/README.md @@ -51,6 +51,21 @@ For complete documentation, visit: [github.com](https://github.com/SYSTRAN/faster-whisper?tab=readme-ov-file#gpu) + +## Info About Speaker Diarization (Detect Different Speakers) + +Set ASR_ENGINE=whisperx in .env-file + +A hugging_face account and a token have to be created (https://huggingface.co/settings/tokens) + +You need to get permission for two models: + +[Speaker Diarization](https://huggingface.co/pyannote/speaker-diarization-3.1) + +[Segmentation]([Segmentation](https://huggingface.co/pyannote/segmentation-3.0)) + +Also see **Request URL Query Params** in this README + ## Credits - This software uses libraries from the [FFmpeg](http://ffmpeg.org) project under the [LGPLv2.1](http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html) -- GitLab