Feature Requests

We have much better results using Open AI for international languages like portuguese, spanish, etc.

Retell has a cheap option of batch calling which costs 0.005 per dial. Please add it otherwise, gets very very expensive.

Ability to modify speed of generated speech - micro changes - i.e. 5% increments to +/- 15%.

Make.com is a very cost effective and powerful competitor to zappier. A Vapi.ai module in Make.com will be amazing!

Enabled the deepgram filler words config: https://developers.deepgram.com/docs/filler-words

Dashboard only displays the 100 newest assistants. We are unable to access assistants that we have developed purely because the dashboard won't scroll up to the past assistants we built, nor can we search for them. Major AI platforms such as OpenAI or ElevenLabs do not leave us with this problem. Having to use an API just to navigate the dashboard does not work given we all work within the VAPI dashboard.

Please add the new Nova-3 model for Deepgram, they boast a much better multilingual support!

The current state of the smart endpointing model is not as great and sometimes just delays the response by almost 1.5 secs when quick words like "Yes" or "Yeah" are spoken. I think adding a more advanced endpointing and turn detection model models are a big request by the community right and would really improve the calls quality. Livekit has implemented a perfect open source plugin that does this for it's text-to-speech models : https://github.com/livekit/agents/blob/main/livekit-plugins/livekit-plugins-turn-detector https://blog.livekit.io/using-a-transformer-to-improve-end-of-turn-detection/ I hope we get something similar or an improved version of the Smart Endpointing model we have :)

VAPI lets you integrate numbers from Twilio and Vonage; Zadarma is just as good as these two (...maybe better) and also requires just an API key and a secret API key like Vonage. It would be really nice to be able to integrate it easily because their offer is much more attractive.

ElevenLabs released a STT model that supports 99 languages. Can you please add it?