Skip to content

Feature Request: Add SenseVoice for on-device voice recognition #61

Description

@LauraGPT

Feature Request

Owl is a great personal wearable AI that runs locally. Suggesting SenseVoice as an ASR option — its compact size and speed make it ideal for local/edge deployment.

Why SenseVoice for wearable AI?

  • 234M params — compact enough for local inference on consumer hardware
  • Non-autoregressive — constant low latency (~100ms), no waiting for sequential decoding
  • 5x faster than Whisper — better battery life and responsiveness
  • 50+ languages — single model handles any language
  • Emotion detection — contextual awareness of speaker mood
  • Audio events — detects environmental sounds (laughter, music, etc.)

Local deployment options

  1. Python (simplest): pip install funasr
  2. ONNX (optimized): via Sherpa-ONNX — supports ARM, x86, mobile
  3. C++ (embedded): SenseVoice.cpp — no Python dependency

Quick start

from funasr import AutoModel
model = AutoModel(model="iic/SenseVoiceSmall", vad_model="fsmn-vad")
result = model.generate(input=audio)

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions