Voices · kalpa-tts-1
Every voice is a person, caught mid-job.
Eleven voices in English and Hindi. Tap or click a voice to hear it. The roles are a suggestion: any voice can take any role.
The cast
Small person, very big chair, all the time in the world.
0:04
Jogging backwards so she can see you.
0:04
Two mugs: one is for the guest.
0:04
Holding an umbrella for someone just out of frame.
0:04
A cup for you, and the dog already likes you.
0:03
Spinning in her chair, laughing at her own caller.
0:05
On tiptoe at the board, chalk on her skirt.
0:03
Four parcels, one eye on the tracking.
0:03
Calm voice, stress ball doing the work.
0:05
Waters the plant between calls.
0:04
That ball is not coming back.
0:04
LiveIn the API
Any of them, by id.
List the voices with GET /v1/voices, then speak in one with POST /v1/tts/{voice_id}. The same voice, every request.
POST/v1/tts/{voice_id}
curl https://api.kalpalabs.ai/v1/tts/dda5319f-3b81-4124-b69e-9241eb510b77 \ -H "Authorization: Bearer $KALPA_API_KEY" \ -H "Content-Type: application/json" \ -d '{"text": "Thanks for holding. I can see your ticket, so let's fix this together."}'# Wren, support