[Feature]: Add Gnani AI Prisma v2.5/v3.0 as a Streaming STT Provider #997
ANAMASGARD
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Overview
I'd like to propose adding Gnani AI Prisma v2.5 and v3.0 as native Speech-to-Text (STT) providers in Dograh, using the official
pipecat-gnaniintegration.Why is this valuable for Dograh?
This would be particularly valuable for Indian customer support, banking, healthcare, and multilingual voice agents.
Proposed Implementation
Integrate both Prisma models into Dograh's existing provider registry and service factory, with model selection, language configuration, and BYOK support.
The official Pipecat plugin already provides streaming STT, so no custom WebSocket implementation is needed.
Initial verification: I successfully tested Prisma v2.5 HTTP STT and Prisma v3.0 HTTP/WebSocket STT against Dograh's Pipecat fork. Full Dograh workflow testing remains to be done.
Official Links
Maintainer Feedback
Would the maintainers be interested in adding Gnani AI as an STT provider?
I'd be happy to implement this feature and raise a PR if approved.
Thank you!

All reactions