1 paper
Patrick S. Burke, Satyam Raj, Sean Kinahan
We present a real-time pipeline of open source, pretrained models for streaming identification of a target speaker, motivated by hearing-aid applications where latency as low as 10…