sherpa-onnx-core
Core shared libraries for sherpa-onnx
Decision gist · record as of 2026-08-14
Yes, if you need local, offline speech or audio AI inference. The package is actively maintained, has no Python dependencies, supports a wide range of platforms and architectures, and carries a permissive license. Install friction is moderate due to platform-specific wheels, but pre-built binaries eliminate the need to compile ONNX Runtime yourself. Verify that your target platform is in the supported list before committing.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires a compatible platform wheel; installation may fail on unsupported architectures (e.g., 32-bit x86 on macOS, or RISC-V on Windows).
- Medium install friction due to platform-specific wheels (9 variants covering macOS, Linux, Windows, and ARM architectures).
- Active maintenance with recent releases (3 days since last version); repository shows strong community engagement with 14181 stars.
License · maintenance · safety
Apache-2.0 (permissive) — Licensed under Apache-2.0 (permissive), allowing free use, modification, and distribution with minimal restrictions—suitable for both open-source and commercial projects.
last release 2026-08-11 (3 days) · last repo commit 2026-08-13 · 14,181 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 526,448 downloads/mo, #6,176 on PyPI
Alternatives
Verify before relying
pip install sherpa-onnx-core
import sherpa_onnx
# Use sherpa_onnx APIs for speech recognition, TTS, or other audio tasks- Whether Python version constraints exist beyond what the fact sheet specifies
- Whether pre-trained models are bundled or must be downloaded separately
- Performance characteristics and latency on different hardware targets
What it is and what it does
sherpa-onnx-core is a Python wrapper around pre-compiled ONNX runtime binaries for running speech and audio AI models locally. It bundles the core C++ libraries needed to execute speech recognition (streaming and non-streaming), text-to-speech, speaker diarization, voice activity detection, keyword spotting, audio tagging, and speech enhancement—all without external service dependencies.
The package targets embedded and edge devices as well as desktop environments, supporting x86, ARM (32 and 64-bit), RISC-V, and specialized NPUs (Rockchip, Qualcomm, Ascend). It has no Python runtime dependencies, only platform-specific binary wheels, making it lightweight for deployment. Users typically pair it with pre-trained ONNX models from the sherpa-onnx ecosystem to perform inference.
Use it for
- Build offline speech-to-text applications on mobile, embedded Linux, or desktop without cloud API calls
- Implement real-time voice activity detection and keyword spotting for always-on voice interfaces
- Deploy speaker identification or diarization in privacy-sensitive environments where audio cannot leave the device
- Add text-to-speech synthesis to applications running on Raspberry Pi, Jetson, or other edge hardware
- Integrate speech enhancement or source separation into audio processing pipelines on resource-constrained devices
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need local, offline speech or audio AI inference.
The package is actively maintained, has no Python dependencies, supports a wide range of platforms and architectures, and carries a permissive license. Install friction is moderate due to platform-specific wheels, but pre-built binaries eliminate the need to compile ONNX Runtime yourself. Verify that your target platform is in the supported list before committing.
Install
sherpa-onnx-core on PyPI
Before you install
Medium install friction due to platform-specific wheels (9 variants covering macOS, Linux, Windows, and ARM architectures). Active maintenance with recent releases (3 days since last version); repository shows strong community engagement with 14181 stars.
Requires a compatible platform wheel; installation may fail on unsupported architectures (e.g., 32-bit x86 on macOS, or RISC-V on Windows).
License in practice
Licensed under Apache-2.0 (permissive), allowing free use, modification, and distribution with minimal restrictions—suitable for both open-source and commercial projects.
Quickstart
pip install sherpa-onnx-core
import sherpa_onnx
# Use sherpa_onnx APIs for speech recognition, TTS, or other audio tasksVerify before relying
- Whether Python version constraints exist beyond what the fact sheet specifies
- Whether pre-trained models are bundled or must be downloaded separately
- Performance characteristics and latency on different hardware targets
Package facts
| License | Apache-2.0 permissive |
| Python support | Not specified |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | None |
| Maintenance | Actively maintained 3 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 526,448 / month, #6,176 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Operating System :: MacOS :: MacOS XOperating System :: Microsoft :: WindowsOperating System :: POSIX :: LinuxProgramming Language :: Python :: 3Topic :: Scientific/Engineering :: Artificial Intelligence |
Evidence: sherpa_onnx_core-1.13.5-py3-none-macosx_10_15_universal2.whl; sherpa_onnx_core-1.13.5-py3-none-macosx_10_15_x86_64.whl; sherpa_onnx_core-1.13.5-py3-none-macosx_11_0_arm64.whl; sherpa_onnx_core-1.13.5-py3-none-manylinux2014_aarch64.whl; sherpa_onnx_core-1.13.5-py3-none-manylinux2014_x86_64.whl; sherpa_onnx_core-1.13.5-py3-none-manylinux_2_35_armv7l.whl; sherpa_onnx_core-1.13.5-py3-none-win32.whl; sherpa_onnx_core-1.13.5-py3-none-win_amd64.whl; sherpa_onnx_core-1.13.5-py3-none-win_arm64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “audio processing onnx”
- sherpa-onnx-coresherpa-onnx-core provides pre-built ONNX runtime binaries for speech…
- sherpa-onnxSherpa-onnx runs speech recognition, text-to-speech, speaker…
- silero-vadSilero VAD detects speech activity in audio files and streams,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also sherpa-onnx · onnx-asr · funasr · pyannote-core · pocketsphinx · SpeechRecognition · speechbrain · pyannoteai-sdk · panns-inference · silero-vad