These examples cover the basic pyannote-style workflows:

• run diarization on one WAV file
• iterate speaker turns
• compute speaker airtime
• reconcile transcript chunks with diarization output

All examples expect:

• a models directory containing segmentation-3.0.onnx, wespeaker-voxceleb-resnet34.onnx, and the PLDA .npy files
• a mono 16kHz 16-bit PCM WAV input

Populate fixtures/models with:

just export-models

cargo run --example diarize_wav -- fixtures/models fixtures/test.wav

Optional file id:

cargo run --example diarize_wav -- fixtures/models fixtures/test.wav meeting-42

This matches pyannote's for turn, speaker in output.speaker_diarization pattern.

cargo run --example print_turns -- fixtures/models fixtures/test.wav

Output:

start   end     speaker
4.705   46.690  SPEAKER_00
...

Useful if you want per-speaker totals.

cargo run --example speaker_airtime -- fixtures/models fixtures/test.wav

Output:

speaker     total_seconds
SPEAKER_00  341.245
...

This example uses exclusive diarization to assign one speaker to each transcript chunk.

Transcript input is a tab-separated file with:

start_seconds<TAB>end_seconds<TAB>text

Example transcript:

0.000	3.200	Welcome everyone
3.200	7.800	Let's get started

Run it with:

cargo run --example assign_transcript_speakers -- fixtures/models fixtures/test.wav transcript.tsv

Output:

start   end     speaker     text
0.000   3.200   SPEAKER_00  Welcome everyone
3.200   7.800   SPEAKER_01  Let's get started