SECONDWAVE TECHNOLOGIES INC. / SAMPLE DELIVERY

This is a fictional demonstration with two neural text-to-speech voices.
It is not client work or a benchmark of transcription accuracy.
The transcript is script-derived. Timestamps are generated speaker-turn boundaries,
not independently measured word-level alignment. Captions use turn-level cues.

Audio: Kokoro-82M v1.0 (Apache 2.0 model), via kokoro-onnx.
Maya: af_heart. Jordan: am_michael. Fictional characters, not real client speakers.
Model: https://huggingface.co/hexgrad/Kokoro-82M
Runtime: https://github.com/thewh1teagle/kokoro-onnx

Files: demo_episode.mp3, transcript.pdf, transcript.txt, subtitles.srt,
subtitles.vtt, episode_index.csv, show_notes.md, demo_segments.json.
The website pairs the player with the full transcript. Click a timestamp to seek.
