Skip to content
Live
Trending storyPast story

PyTorch optimizing Arm® edge device streaming speech recognition

1 reports1 outlets23 days agoUpdate

The story so far

AI overview

PyTorch optimizes streaming speech recognition on Arm® edge devices through ExecuTorch, with Nemotron 3.5 achieving a maximum accuracy of 11.7% WER on Google Pixel 11, 3.5x faster than FP32, but with increased latency; short audio chunks can reduce latency, but accuracy drops to 15.2% WER.

AI-written from the reports · Updated 23 days ago

Coverage timeline

Follow the coverage to see every side of the story.

27 Oct
  1. PyTorch Blog
    Optimizing Streaming Speech Recognition on Arm® Edge Devices with ExecuTorch

    Nemotron 3.5 achieved the highest streaming speech recognition accuracy on Google Pixel 11 using ExecuTorch and SME2, reaching up to 11.7% WER (with 1,120 ms audio blocks and QC8W precision), which is 3.5x faster than FP32 but with increased latency; shorter audio blocks (e.g., 80 ms) can reduce latency but accuracy drops to 15.2% WER.