Real-time speech recognition and voice activity detection (VAD) using next-gen Kaldi with ncnn without Internet connection. Support iOS, Android, Linux, macOS, Windows, Raspberry Pi, VisionFive2, LicheePi4A etc.
APACHE-2.0 License
Real-time Speech recognition | Voice activity detection |
---|---|
✔️ | ✔️ |
Architecture | Android | iOS | Windows | macOS | linux |
---|---|---|---|---|---|
x64 | ✔️ | ✔️ | ✔️ | ✔️ | |
x86 | ✔️ | ✔️ | |||
arm64 | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ |
arm32 | ✔️ | ✔️ | |||
riscv64 | ✔️ |
1. C++ | 2. C | 3. Python | 4. JavaScript |
---|---|---|---|
✔️ | ✔️ | ✔️ | ✔️ |
5. Go | 6. C# | 7. Kotlin | 8. Swift |
---|---|---|---|
✔️ | ✔️ | ✔️ | ✔️ |
It also supports WebAssembly.
This repository supports running the following functions locally
on the following platforms and operating systems:
x86_64
, 32-bit ARM, 64-bit ARM (arm64, aarch64), RISC-V (riscv64)with the following APIs
C#
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.,
We provide a few YouTube videos for demonstration about real-time speech recognition
with sherpa-ncnn
using a microphone:
Multilingual (Chinese + English) with endpointing Python demo : https://www.bilibili.com/video/BV1eK411y788/
Android demos
Multilingual (Chinese + English) Android demo 1: https://www.bilibili.com/video/BV1Ge411A7XS
Multilingual (Chinese + English) Android demo 2: https://www.bilibili.com/video/BV1eK411y788/
Chinese (with background noise)
Android demo : https://www.bilibili.com/video/BV1GR4y167fx
Chinese
Android demo : https://www.bilibili.com/video/BV1744y1Z76H
Chinese poem with background music
Android demo : https://www.bilibili.com/video/BV1vR4y1k7eo
Description | URL |
---|---|
Streaming speech recognition | Address |
https://github.com/k2-fsa/sherpa-ncnn/releases/tag/models
Please see https://k2-fsa.github.io/sherpa/social-groups.html for 新一代 Kaldi 微信交流群 and QQ 交流群.