How Have Users of 11Labs Text-to-Speech Software Managed the Long Wait Time in Synchronous Process within Pipe Dream Workflow?

Help
1

This topic was automatically generated from Slack. You can find the original thread here.

Hi, for those who have used the 11Labs text-to-speech software, how have you handled the long wait time for a synchronous process, which makes the Pipe Dream workflow expensive? I’m considering moving it out of Pipe Dream as a step, but before I go through that, I wanted to hear what other people think and have done about this.

2

Is there a way to make it async in 11Labs?

3

Hi – You can pause/resume workflows. That is the standard approach when executing an API request that is long running: GitHub

Going off what Andrew was saying, ideally the 3rd party API emits an event when the process is complete on their side.

4

I think that’s what Andrew is asking about — ideally the action would handle that by kicking off an async process w/ 11Labs, then the action could suspend execution and either either poll for completion or wait for a callback to resume.

5

I see that’s pretty useful to know that you can pause and resume workflows. I didn’t realize we could do that.

I ended up making it asynchronous. I created a Docker process in Google Cloud Run and ran it that way. It issued an immediate response back to Pipedream so it wouldn’t wait for job completion.