undefined | Better HN

0 pointsranger_danger4d ago0 comments

How is this so much faster than even GPU-based whisper?

0 comments

small, ONNX-optimized models designed specifically for low-latency CPU streaming, so it avoids overhead of large transformer arch and GPU memory transfers

j / k navigate · click thread line to collapse

0 comments

mrkn14d ago

small, ONNX-optimized models designed specifically for low-latency CPU streaming, so it avoids overhead of large transformer arch and GPU memory transfers

j / k navigate · click thread line to collapse