Show HN: Outspeed – Platform for realtime voice and video AI

[jankboy]

Hey HN! Janak here from Outspeed (https://outspeed.com).

We’re excited to show you Outspeed : a purpose-built platform for realtime voice & video AI applications.

Here’s a demo of some cool apps you can create using Outspeed: https://www.youtube.com/watch?v=a11LQIlXelM

Outspeed emerged from our frustration of needing to stitch together multiple tools such as livekit, vocode, langflow, silero etc. just to make a simple voice bot. Even after all that hard work, it still wasn’t production-ready. So we decided to work on a complete framework that could stand production level workloads.

Outspeed differs from other open-source libraries such as Pipecat or Livekit-Agents in 3 major ways:

1. Pytorch-like interface - Livekit and Pipecat were built on video-conferencing primitives and thus, are non-intuitive for a python/ML developer.

2. Vercel-like deployments - You can deploy your code using a single command to Outspeed’s cloud or host it on your own infra.

3. Built-in WebRTC server - Instead of deploying another server to handle webRTC connections, Outspeed comes with a built-in webRTC server. No longer need to depend on webRTC providers such as Livekit or Daily.

Outspeed is being actively developed. We’re eager to hear honest feedback, likes, dislikes, feature requests, you name it.

[shravan]

Congratulations on the launch! I'm curious to understand, in your experience, what was the most challenging part of building a realtime voice AI app was? Naively, assumed that this would be a solved problem.

[jankboy]

The hardest part has to be building webRTC streaming support from scratch into our SDK to provide extremely low latency API. We had this idea where we imagined the webRTC server and AI model running on the same instance and talking to the user via a direct peer connection. As far as we know, this architecture is first of its kind and isn't usually how webRTC peer connections are made.

[saharshagrawal]

Interesting library, is it like AWS lambda for realtime AI? How does the serverless offering work?

[jankboy]

That's right but instead we built on top of audio-video streaming primitives so that it provides the same convenience of AWS lambda but can support persistent webRTC & websocket connections. All you have to do is write your backend logic using our open source SDK and then you can one click deploy to Outspeed's infra similar to Vercel.