Нема описа

Fangjun Kuang 2e89994fc6 Fix building. (#325) пре 1 година
.github 2e89994fc6 Fix building. (#325) пре 1 година
android ca9c19facd Add Android WearOS demo (#319) пре 1 година
c-api-examples 401de81194 Add hotwords support to Swift/Go/Python/C#/Kotlin APIs (#260) пре 1 година
cmake 884ce6d9eb Use tencent/ncnn master. (#320) пре 1 година
dotnet-examples 77c34f3cff Add microphone example for .Net (#248) пре 2 година
ffmpeg-examples 0ba47330ce FFmpeg: Feed PCM to K2 direclty, without any buffer. (#183) пре 2 година
go-api-examples 343eab32ab Update sherpa-ncnn-go version (#245) пре 2 година
ios-swift b4260f7acf Refactor code (#163) пре 2 година
ios-swiftui f12aac47c3 Add SwiftUI demo project (#175) пре 2 година
mfc-examples 98ba3218db Fix MFC examples (#268) пре 1 година
nodejs-examples 41ebcfb11b Add WebAssembly + NodeJS example. (#311) пре 1 година
python-api-examples 87899c26fe Add Python ASR example with alsa (#324) пре 1 година
scripts 2e89994fc6 Fix building. (#325) пре 1 година
sherpa-ncnn 2e89994fc6 Fix building. (#325) пре 1 година
swift-api-examples 401de81194 Add hotwords support to Swift/Go/Python/C#/Kotlin APIs (#260) пре 1 година
toolchains f9a850c725 Support RISC-V (#213) пре 2 година
wasm 41ebcfb11b Add WebAssembly + NodeJS example. (#311) пре 1 година
.clang-format 78ce6f6233 Add CI test and refactor the code a bit. (#4) пре 2 година
.gitignore 41ebcfb11b Add WebAssembly + NodeJS example. (#311) пре 1 година
CMakeLists.txt 87899c26fe Add Python ASR example with alsa (#324) пре 1 година
CPPLINT.cfg 78ce6f6233 Add CI test and refactor the code a bit. (#4) пре 2 година
LICENSE f91cd77f83 Use standard apache 2.0 license (#123) пре 2 година
MANIFEST.in 2b5300e975 Add Python API (#47) пре 2 година
README.md 2e78bbf587 Support endpoint detection in Python (#56) пре 2 година
build-aarch64-linux-gnu.sh 533a44c25a Support windows x86 (#84) пре 2 година
build-android-arm64-v8a-with-vulkan.sh 6b7072b3d9 Pack all sources into a single tar for embedded systems (#96) пре 2 година
build-android-arm64-v8a.sh 6b7072b3d9 Pack all sources into a single tar for embedded systems (#96) пре 2 година
build-android-armv7-eabi.sh 6b7072b3d9 Pack all sources into a single tar for embedded systems (#96) пре 2 година
build-android-x86-64.sh 6b7072b3d9 Pack all sources into a single tar for embedded systems (#96) пре 2 година
build-android-x86.sh 2878e796a5 build script for android-x86 (#228) пре 2 година
build-apk.sh fc4d3bc2d9 Rlease pre-built APK for English and French (#256) пре 1 година
build-arm-linux-gnueabihf.sh 6c8cff9580 Disble portaudio for cross-compiling (#79) пре 2 година
build-ios.sh 7a5f80edc4 fix building for iOS (#92) пре 2 година
build-m3axpi.sh b701bc8daa Fix building on m3axpi (#99) пре 2 година
build-riscv64-linux-gnu.sh f9a850c725 Support RISC-V (#213) пре 2 година
build-swift-macos.sh da14875cbb Refactor decoding (#132) пре 2 година
build-wasm-simd-for-nodejs.sh 41ebcfb11b Add WebAssembly + NodeJS example. (#311) пре 1 година
build-wasm-simd.sh 72ea103e9b Add WebAssembly (#300) пре 1 година
install-vulkan-macos.md d5dd41a7e2 Support building android app (#37) пре 2 година
pack-for-embedded-systems.sh 884ce6d9eb Use tencent/ncnn master. (#320) пре 1 година
release.sh 797f1e4243 Release pre-compiled libs for Android and iOS (#181) пре 2 година
setup.py df9ae1ff84 Publish to PyPI (#48) пре 2 година

README.md

Introduction

You can use sherpa-ncnn for real-time speech recognition (i.e., speech-to-text) on

  • Linux
  • macOS
  • Windows
  • Embedded Linux (32-bit arm and 64-bit aarch64)
  • Android
  • etc ...

We support all platforms that ncnn supports.

Everything can be compiled from source with static link. The generated executable depends only on system libraries.

HINT: It does not depend on PyTorch or any other inference frameworks other than ncnn.

Please see the documentation https://k2-fsa.github.io/sherpa/ncnn/index.html for installation and usages, e.g.,

  • How to build an Android app
  • How to download and use pre-trained models

We provide a few YouTube videos for demonstration about real-time speech recognition with sherpa-ncnn using a microphone:

See also https://github.com/k2-fsa/sherpa