Mark Backman
Mark Backman
@pesterhazy I haven't seen this with Cartesia. I'm wonder if there's something specific to your pipeline. Would you mind sharing the services you're using for this? Also, is this in...
@pesterhazy is this still an issue?
@hdcaruth do you have a reproducible repro case? If so, do you mind sharing? If not, how frequently does this occur for you? Can you also share: - what services...
Pipecat is used primarily in real-time use cases. Is this real-time video generation? If not, how would you expect these to work in the context of Pipecat?
I've notified the Speechmatics team about this. They maintain the integration and are planning to look into this soon.
We are definitely interested in this but time hasn't permitted just yet. If anyone is interested, we're happy to collaborate on this.
@shubhamofbce we aren't working on this. LLMs in Pipecat are quite complex and require a good understanding of how Pipecat works. If you decide to work on this, please look...
We have discussed adding a WebsocketTransport to the client SDKs. This is something we plan to do in the near future. Just an FYI: sending audio and video to a...
There are no known memory leaks in Pipecat. To look into this, you'll need to provide more detailed information on what you're doing and what you're seeing.
I have tested with Deepgram over thousands of calls and minutes and I've never seen this issue. If you have a repro case, please share.