Sarvam AI makes speech models built for Indian languages. Choose it when your agent talks to callers in Hindi or other Indian languages and you want a natural Indian voice.
ttsProvider |
sarvam |
| Voices | 43, multilingual |
| Example voices | kavitha, ritu, priya, neha, pooja, simran |
The full voice list is in the Provider & Voice Catalog.
Set it in the dashboard#
- Open your agent at voice-agent.edesy.in.
- In the voice settings, pick Sarvam AI as the text-to-speech provider.
- Choose a voice and place a test call to hear it.
Set it with the API#
curl -X PATCH https://voice-agent.edesy.in/api/v1/agents/{agentId} \
-H "Authorization: Bearer $EDESY_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "ttsProvider": "sarvam", "ttsVoice": "kavitha" }'
Only the fields you send change. See Agents API.
Note Sarvam also offers speech-to-text, with a different id: sarvam_stt. Use sarvam only in ttsProvider and sarvam_stt only in sttProvider. A TTS id in sttProvider is accepted when you save but fails at call time.