Skip to content
pubscout

sherpa_onnx

2.0.2, released 2 days ago by an unverified publisher

On‑device Flutter library for speech recognition, synthesis, speaker and audio tagging functions.

$ flutter pub add sherpa_onnx
Source repository

What it does

Provides offline APIs for speech‑to‑text, text‑to‑speech, speaker identification, audio tagging, VAD and other audio AI tasks. It runs on multiple platforms via native ONNX runtimes.

AI-written summary of version 2.0.1, based on the package's README and metadata. Report an inaccuracy

Use it when

  • You need speech or audio AI features without internet connectivity
  • Your app targets Android, iOS, web, desktop and wants a single Flutter package
  • You want support for many audio functions (ASR, TTS, speaker ID, VAD, etc.)

Skip it when

  • You require cloud‑based large language models or streaming from a server
  • App size must be minimal and native binaries are a concern
  • You need only simple audio playback without AI processing

Strengths

  • Fully offline operation
  • Broad set of audio AI capabilities
  • Cross‑platform support including web and desktop
  • Supports hardware NPUs for acceleration

Weaknesses

  • Large native binaries increase app size
  • Setup involves platform‑specific native libraries and permissions
  • Documentation for Flutter integration is limited

Setup

Kind
Plugin (native code)
Effort
Medium
Code generation
No
Direct dependencies
10

Add the sherpa_onnx package and platform‑specific sherpa_onnx_* dependencies, configure ffi, grant microphone permissions, and ensure the correct native binaries are bundled for each target platform.

Alternatives

Packages that do the same job and could replace sherpa_onnx.

Works with

Packages commonly used together with sherpa_onnx: ones it builds on, ones built on it, and its own family.