AWS transcribe aws-sdk-cpp above 1.11.283 version CPU spikes more then 100%

[vdharashive]

Describe the bug

Reported issue in below thread is not resolved

#2779

Expected Behavior

Performance issue with asr aws transcribe

Current Behavior

CPU usage for each connection goes avg 80%

Reproduction Steps

yes its reproducible

Possible Solution

No response

Additional Information/Context

No response

AWS CPP SDK version used

1.11.283

Compiler and Version used

gcc (Debian 12.2.0-14) 12.2.0

Operating System and version

Debian GNU/Linux 12 (bookworm)

[vdharashive]

https://github.com/jambonz/freeswitch-modules/tree/main/mod_aws_transcribe

is code implementation

@davehorton

[jmklix]

We are currently looking into this.

[jmklix]

We added an additional configuration to help avoid cpu spikes when using transcribe. Please update to the latest version of this sdk and let us know if you have any other questions/concerns when using this sdk.

[vdharashive]

@jmklix will test and let you know thank you very much

[vdharashive]

@jmklix having LOW_LATENCY and REGULAR is not ideal solution , since for real-time transcript with REGUALR does not respond quick enough and gets stuck, with LOW_LATENCY cpu usage goes very high. i don't think is right way to address the issue

does using websocket api will solve this problem ?

[jmklix]

Can you elaborate on what you exactly mean with "REGUALR does not respond quick enough and gets stuck"? How quickly are you expecting this sdk to respond?

What are your thoughts/suggestions on the correct way to fix this issue? How would you have websocket api to improve this?

[github-actions]

Greetings! It looks like this issue hasn’t been active in longer than a week. We encourage you to check if this is still an issue in the latest release. Because it has been longer than a week since the last update on this, and in the absence of more information, we will be closing this issue soon. If you find that this is still a problem, please feel free to provide a comment or add an upvote to prevent automatic closure, or if the issue is already closed, please feel free to open a new one.