Speech, one POST away.
Conversational text-to-speech on dedicated H100s. Realtime synthesis, billed by the second of audio. Nothing else.
curl heycollin.dev/v1/tts -H "Authorization: Bearer ck_live_…" -d '{"text":"Hey."}'
Free demo: 3 clips. Sign in for 10 minutes, a playground and the API · every clip is watermarked