PlayHT is building an infrastructure for Conversational AI Voice Agents so every business, developer, or tinkerer can easily build talking human-like AI agents and use them to serve their customers. To do that PlayHT built the fastest ever realtime generative text to speech models, which you can use at scale with any LLM seamlessly.
Here is a demo of PlayHT's API integrated with chatGPT with input and output streaming to achieve the lowest latency, check this demo's code here.
It's super easy to get started. First thing you need are your credentials - generate an API Secret Key and get your User ID.
Use PlayHT API for -
- Realtime Text to Speech streaming.
- Async long form Text to Speech generation.
- Ultra realistic Instant Voice Cloning.
Our API supports Input streaming to make it seamless to integrate with LLMs like chatGPT, check this guide to see an example of how to use it.
Check this guide for some techniques to keep in mind when trying to achieve the lowest latency.
Follow this guide for integrating realtime audio streaming with Twilio for over the phone conversations.
For non-streaming generation of long form content where you need a full audio file and not a realtime buffer stream, get started with this guide.
To get more info about the supported plans, please visit https://play.ht/pricing/ and select the "API" tab.
With higher plans you can get higher concurrency rate limits, a performance SLA and better support.