sci-ml/sherpa-onnx (bentoo)

Search

Install

Install this package:

emerge -a sci-ml/sherpa-onnx

Package Information

Description:
sherpa-onnx is the speech toolkit of the k2-fsa project: speech-to-text, text-to-speech, speaker diarization, voice activity detection, source separation and keyword spotting, all running on ONNX Runtime with no PyTorch dependency. This is a source build linked against the system sci-libs/onnxruntime rather than a bundled copy, which is what SHERPA_ONNX_USE_PRE_INSTALLED_ONNXRUNTIME_IF_AVAILABLE selects. Upstream's CMake build resolves roughly a dozen small dependencies through FetchContent at configure time — eigen, asio, cargs, json, kaldi-decoder, kaldi-native-fbank, kaldifst, openfst, kissfft, simple-sentencepiece, hclust-cpp, and optionally espeak-ng, piper-phonemize, portaudio, websocketpp and pybind11. This ebuild declares every one of them in SRC_URI, pinned by tag or full commit SHA, and stages them into the source directory so CMake resolves them locally. The build therefore needs no network access and carries no RESTRICT="network-sandbox". Model files are not bundled; each task loads its own from upstream. See https://k2-fsa.github.io/sherpa/onnx/pretrained_models/
Homepage:
https://k2-fsa.github.io/sherpa/onnx/ https://github.com/k2-fsa/sherpa-onnx
License:
Apache-2.0

Versions

Version EAPI Keywords Slot
1.13.6 8 ~amd64 0

Metadata

Description

Maintainers

Upstream

Raw Metadata XML
<pkgmetadata>
	<maintainer type="person">
		<email>lucascs@proton.me</email>
		<name>Lucas C.S.</name>
	</maintainer>
	<longdescription lang="en">
		sherpa-onnx is the speech toolkit of the k2-fsa project:
		speech-to-text, text-to-speech, speaker diarization, voice activity
		detection, source separation and keyword spotting, all running on ONNX
		Runtime with no PyTorch dependency.

		This is a source build linked against the system sci-libs/onnxruntime
		rather than a bundled copy, which is what
		SHERPA_ONNX_USE_PRE_INSTALLED_ONNXRUNTIME_IF_AVAILABLE selects.

		Upstream's CMake build resolves roughly a dozen small dependencies
		through FetchContent at configure time — eigen, asio, cargs, json,
		kaldi-decoder, kaldi-native-fbank, kaldifst, openfst, kissfft,
		simple-sentencepiece, hclust-cpp, and optionally espeak-ng,
		piper-phonemize, portaudio, websocketpp and pybind11. This ebuild
		declares every one of them in SRC_URI, pinned by tag or full commit
		SHA, and stages them into the source directory so CMake resolves them
		locally. The build therefore needs no network access and carries no
		RESTRICT="network-sandbox".

		Model files are not bundled; each task loads its own from upstream.
		See https://k2-fsa.github.io/sherpa/onnx/pretrained_models/
	</longdescription>
	<use>
		<flag name="cuda">Build the NVIDIA CUDA execution provider (requires the CUDA Toolkit; the nvcc host compiler is resolved from the installed toolkit, not hardcoded)</flag>
		<flag name="portaudio">Build the example CLI tools that record and play audio. Adds no system dependency: upstream vendors and statically links portaudio</flag>
		<flag name="python">Build and install the Python bindings (pybind11) into site-packages</flag>
		<flag name="tts">Build text-to-speech support, which pulls the vendored espeak-ng and piper-phonemize sources</flag>
		<flag name="websocket">Build the WebSocket server and client examples</flag>
	</use>
	<upstream>
		<bugs-to>https://github.com/k2-fsa/sherpa-onnx/issues</bugs-to>
		<remote-id type="github">k2-fsa/sherpa-onnx</remote-id>
	</upstream>
</pkgmetadata>

Lint Warnings

USE Flags

Manage flags for this package: euse -i <flag> -p sci-ml/sherpa-onnx | euse -E <flag> -p sci-ml/sherpa-onnx | euse -D <flag> -p sci-ml/sherpa-onnx

Flag Description 1.13.6
cuda Build the NVIDIA CUDA execution provider (requires the CUDA Toolkit; the nvcc host compiler is resolved from the installed toolkit, not hardcoded)
portaudio Build the example CLI tools that record and play audio. Adds no system dependency: upstream vendors and statically links portaudio
python Build and install the Python bindings (pybind11) into site-packages
tts Build text-to-speech support, which pulls the vendored espeak-ng and piper-phonemize sources
websocket Build the WebSocket server and client examples

Manifest

Type File Size Versions
DIST asio-asio-1-24-0.tar.gz 2902485 bytes 1.13.6
DIST cargs-1.0.3.tar.gz 41253 bytes 1.13.6
DIST eigen-5.0.1.tar.gz 2967272 bytes 1.13.6
DIST espeak-ng-ed530aa113046142eb5115cf2fc9157854d0ffe1.zip 18011501 bytes 1.13.6
DIST hclust-cpp-2026-02-25.tar.gz 22523 bytes 1.13.6
DIST json-3.12.0.tar.gz 9678593 bytes 1.13.6
DIST kaldi-decoder-0.3.0.tar.gz 51199 bytes 1.13.6
DIST kaldi-native-fbank-1.22.3.tar.gz 71144 bytes 1.13.6
DIST kaldifst-1.8.0.tar.gz 172147 bytes 1.13.6
DIST kissfft-febd4caeed32e33ad8b2e0bb5ea77542c40f18ec.zip 74252 bytes 1.13.6
DIST openfst-1.8.5-2026-07-09.tar.gz 1501685 bytes 1.13.6
DIST pa_stable_v190700_20210406.tgz 1462695 bytes 1.13.6
DIST piper-phonemize-f3ff95afc03640bc1399e113e83361192a2fafb4.zip 9805962 bytes 1.13.6
DIST pybind11-3.0.0.tar.gz 915234 bytes 1.13.6
DIST sherpa-onnx-1.13.6.gh.tar.gz 11564933 bytes 1.13.6
DIST simple-sentencepiece-0.7.tar.gz 355335 bytes 1.13.6
DIST websocketpp-b9aeec6eaf3d5610503439b4fae3581d9aff08e8.zip 996004 bytes 1.13.6
Unmatched Entries
Type File Size