@nitesh_singh5 Working on STT/ASR for conversational agents. Experienced with reproducing complex model architectures, working on the inference layer for low latency operations. Prior publication experience as well.
@ishaangodhaa Havent worked with harness much, but presently researching a lot on the inference layer. Specifically working with improving the latency of stt models by various decoding strategies.
P.S targetting less than 200ms
@Atharv_Golait hey,
working in building SOTA models for STT/ASR in the indic voice AI domain. Directly working on the inference layer (decoding strategies) to improve latency! Let's connect :)