Streaming?
#11
by dyqiang - opened
Thank you NVIDIA team for releasing yet another excellent ASR model!
Is there a guide on how to achieve streaming transcription using the latest parakeet-tdt-0.6b-v3 model?
We do also have dedicated cache-aware architecture for streaming use cases: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/stt_en_fastconformer_hybrid_large_streaming_multi . We are also working on an upgraded performant model to this one.
I wonder if the more advanced version has been developed yet? Where can I find relevant information?
Is the cache awareness confirmed to work with this model? @nvidia