If GPU/memory resource is enough, is there any configuration to limit the maximum number of concurrent requests processed in one Riva ASR streaming recognition server instance?
HI @174362510
Thanks for your interest in Riva
I will check regarding this query with Riva team and update you
Thanks
Hello rvinobha,
Is there any update for this issue? It seems that we can config triton related parameters such as PREFERRED_BATCH_SIZE in config.sh under riva_quickstart folder, right? Could you explain it in detail?
Hello @rvinobha , sorry for disturbing you, is there any update for this issue?
Hi @174362510 and @whz796100
Apologies for the delay,
I will check again today with internal team and get back
Thanks