Python implementation of the NIST md-eval.pl script for evaluating rich transcription and speaker diarization accuracy.
Voice-bank-first speaker labelling for the Plaud Note family. Local, macOS Apple Silicon, AGPL-3.0.
An English-Spanish code switching dataset adapted from the Miami-Corpus
Fully automated multi-speaker transcription tool built off of the WhisperX library with Pyannote. Comes with Word Alignment, Timestamps, Voice activity Detection and Speaker Diarization. Mathematically optimised for accuracy.
Transcription audio LOCAL M4A avec diarisation des locuteurs - Solution utilisant faster-whisper et clustering MFCC
A Python CLI tool for transcribing podcasts with automatic speaker diarisation using OpenAI's Whisper API. Features real-time progress tracking, cost estimation, automatic chunking for large files, and clean Markdown output with timestamps. Supports both URLs and local files. Python 3.10+ compatible.
Local-only clinical scribe for macOS. Every sentence in the note cites the words that were said. Rust engine + SwiftUI shell, nothing leaves the Mac. Apache-2.0.
Transcribe interviews and meetings locally, with speaker labels and timestamps (faster-whisper + pyannote.audio)
Transcription et séparation des voix en local sur Mac Apple Silicon, sans compte ni clé d'API. En cours de développement.
To associate your repository with the diarisation topic, visit your repo's landing page and select "manage topics."