skills.rest
jayll1303
jayll1303COMMUNITY
Explore all skills by this author →
GitHub

sherpa-onnx

Perform offline speech processing with ONNX models via sherpa-onnx.

17|3|Updated Apr 3, 2026
Search Tags:#tts#asr#vad#sherpa-onnx#speaker-diarization#onnx-runtime#offline-inference
One-click install
npx skills add https://github.com/jayll1303/AIEKit --skill sherpa-onnx
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sherpa-onnx
Source: https://github.com/jayll1303/AIEKit/tree/main/.kiro/skills/sherpa-onnx
Command: npx skills add https://github.com/jayll1303/AIEKit --skill sherpa-onnx

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Enable deterministic, private, and low-latency speech processing locally without internet access by providing ready guidance to run ASR, TTS, VAD, speaker diarization, speaker ID/verification, speech enhancement, audio tagging, keyword spotting, and source separation using ONNX models and the sherpa-onnx runtime.

Core Features & Use Cases

  • Streaming & Non‑streaming ASR: real-time microphone transcription via OnlineRecognizer and batch/file transcription via OfflineRecognizer.
  • TTS Engines: generate speech locally with Kokoro, Piper, Matcha, VITS, or KittenTTS for multi‑speaker and multi‑language needs.
  • VAD, Diarization & Speaker Tasks: voice activity detection, segmentation, embedding extraction, identification and verification workflows for meetings and call analytics.
  • Enhancement & Tagging: denoise or separate sources, classify audio content, and detect keywords on-device for privacy-sensitive or edge applications.
  • Use Case Example: transcribe a meeting audio file on an offline workstation, split speakers using pyannote segmentation plus embeddings, and export timestamped captions and per-speaker transcripts.

Quick Start

Download the appropriate ONNX model, install sherpa-onnx, and run a local offline transcription of meeting.wav with OfflineRecognizer.from_sense_voice to produce a timestamped transcript.

Frequently Asked Questions about sherpa-onnx

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run offline speech recognition and text-to-speech without an internet connection?▼

Offline speech recognition and text-to-speech require downloading local ONNX models and using a runtime like sherpa-onnx to perform deterministic CPU or CUDA inference without internet access.

Can I perform speaker diarization and voice activity detection locally on edge devices?▼

Yes, speaker diarization and voice activity detection can run locally on edge devices by applying ONNX models to segment audio, extract speaker embeddings, and identify speakers completely offline.

What is the best way to transcribe a meeting audio file and split it by speaker offline?▼

To transcribe meeting audio and split speakers offline, use a non-streaming OfflineRecognizer for timestamped transcription and apply pyannote segmentation with speaker embeddings for per-speaker transcripts.

How do I generate multi-language speech locally using ONNX models?▼

Generate multi-language speech locally by loading TTS ONNX model artifacts such as VITS, Kokoro, or Piper into the sherpa-onnx runtime to produce audio without cloud dependencies.

Does offline inference with ONNX Runtime support both streaming and non-streaming ASR?▼

Yes, ONNX Runtime supports both streaming ASR via OnlineRecognizer for real-time microphone transcription and non-streaming ASR via OfflineRecognizer for batch file processing.

Do I need to download specific model files and tokens to run local audio processing?▼

Yes, local audio processing requires downloading specific ONNX model files, tokens, and voice configurations alongside setting correct sample rates to ensure deterministic offline inference.

Related Skills

Explore More Software Engineering Skills→
fikriaffikriaf

llama-cpp

Automate discovery, download, and local inference of llama.cpp models from Hugging Face Hub.

Community
Advanced
akrindevakrindev

video-transcriber

Transcribe video audio into timestamped SRT, VTT, or JSON transcripts.

Community
Advanced
Canhada-LabsCanhada-Labs
2

voice-ai-integration

Integrate ASR and TTS providers with real-time audio streaming and speaker diarization.

Official
Advanced
NVIDIANVIDIA
2.8k

nemotron-voice-agent-deploy

Deploy Nemotron Voice Agent on Workstation, Jetson Thor, or Cloud NIMs with WebRTC or WebSocket transport.

Official
Advanced