voice

voice — STT → TTS round-trip

Drop a short MP3 in this folder as input.mp3, then:

cp .env.example .env
pnpm install
pnpm start
# transcript: ...
# wrote output.mp3

STT alternatives: deepgram/nova-3, openai/gpt-4o-transcribe. TTS alternatives: openai/tts-1-hd, elevenlabs/turbo-v2.5.