With Bolna, developers can plug in their own LLMs, telephony systems, transcription models, and voice synthesis tools - and rapidly build apps that talk to users over phone calls or browser voice.
Here’s a short demo video: https://github.com/bolna-ai/bolna/assets/1313096/2237f64f-1c...
Why we built this We originally set out to build an AI interviewer for practicing coding interviews - think Leetcode with voice. But we quickly realized that the orchestration layer (LLM ↔ telephony ↔ TTS ↔ ASR ↔ prompts ↔ logic) was the real challenge.
If this sounds interesting to you, please check us out! You can check us out our open source github repository at Github - https://github.com/bolna-ai or our hosted no-code/low-code dashboard at - https://platform.bolna.ai.
We’re inspired by open ecosystems like the Elastic stack and trying to follow a similar path: open by default, but easy to extend or host as needed.