Speaker verification in Jarvis

Hi,
Jarvis Speech Services currently (1.2.0-beta) does not seem to support speaker verification (i.e., extracting embedding from NeMo SpeakerNet models). Is support for this planned? If not, is this something I could add to the Jarvis Speech Services API server myself?
Thanks

Hi @ee_nvidiangc
Speaker verification is not supported currently in Jarvis Speech Services.

Thanks

Hi @SunilJB
Yes, I understand. That’s why I was asking whether there are plans for supporting it in the future, and if not, whether this is something users could add themselves (relatively easily)?
Thanks

Hi @ee_nvidiangc
Could you please share the use case that you are trying to implement in our case?

Thanks

Hi @SunilJB
Thanks for you reply!
The specific use case is biometric authentication of call participants based on voice. A possible scenario is outlined in Integrating with Telephone Networks to Enable Real-Time AI Services | NVIDIA Developer Blog, where Jarvis is used for transcription of phone calls.
If Jarvis supported extracting embeddings from SpeakerNet NeMo models, these embeddings could be used to authenticate call participants (speaker verification).
Please let me know if you’d like me to provide more details.
Thanks