speaker-diarization-community-1

by pyannote

Community speaker diarization pipeline for who-spoke-when audio metadata

2.7Mdl/month

526likes

Pipelineparams

HuggingFace Use in Pipeline

Identifiers

Model ID

pyannote/speaker-diarization-community-1

Feature URI

mixpeek://transcription@v1/pyannote_diarization_community_1

Overview

pyannote Community-1 is a speaker diarization pipeline that segments audio by speaker turns, speech activity, speaker changes, and overlapped speech. It is publicly accessible with license acceptance and has become one of the highest-traffic diarization models on HuggingFace.

On Mixpeek, diarization turns raw audio and video transcripts into searchable conversational structure. Agents can ask not only what was said, but who said it and when it happened.

Architecture

pyannote.audio pipeline composed of voice activity detection, speaker change detection, overlapped speech detection, embedding, and clustering components. Accepts whole files or waveform excerpts.

Mixpeek SDK Integration

import { Mixpeek } from "mixpeek";

const mx = new Mixpeek({ apiKey: "API_KEY" });

await mx.collections.ingest({
  collection_id: "meetings",
  source: { url: "https://example.com/meeting.wav" },
  feature_extractors: [{
    feature: "speaker_diarization",
    model: "pyannote/speaker-diarization-community-1"
  }]
});