Speaker Diarization & Voice Analysis
14 repos
Python libraries and models for identifying and tracking speakers in audio, including speaker diarization (who spoke when), speaker verification (confirming speaker identity), voice activity detection, and speaker embedding extraction. The cluster centers on pyannote-audio as a foundational framework, with specialized model variants (diarizen-wavlm) and complementary tools like diart and wespeaker providing alternative implementations, optimization approaches (ONNX), and production-ready pipelines for real-world audio processing tasks.
Python · 6
Jupyter Notebook · 2
HTML · 1
speaker-diarization ·12,774
voice-activity-detection ·12,715
speaker-embedding ·12,537
pytorch ·10,593
speaker-recognition ·10,525
overlapped-speech-detection ·10,514
speaker-change-detection ·10,514
pretrained-models ·10,514
speaker-verification ·10,514
speech-activity-detection ·10,514
