Hi, This is a fantastic solution for a local high-performant Speech to text capability.
I seek some inputs
1) One stream/client does seem to take decent amount of CPU, I would like to know if there are any broad indicators on how many simultaneous sessions can be supported on CPU / GPU
2) Can it support multiple web clients simultaneously in the current version for multiple ongoing Speech to text sessions.
3) Is there a setting that can auto-detect language and provide the translated result and detected -language? Any reference /inputs to modify or look into specific code would also be very useful.
Hi, This is a fantastic solution for a local high-performant Speech to text capability.
I seek some inputs
1) One stream/client does seem to take decent amount of CPU, I would like to know if there are any broad indicators on how many simultaneous sessions can be supported on CPU / GPU
2) Can it support multiple web clients simultaneously in the current version for multiple ongoing Speech to text sessions.
3) Is there a setting that can auto-detect language and provide the translated result and detected -language? Any reference /inputs to modify or look into specific code would also be very useful.
Thank you. Deepak