https://github.com/skit-ai/kaldi-serve
Server framework for Kaldi ASR Toolkit
https://github.com/skit-ai/kaldi-serve
asr grpc grpc-server kaldi kaldi-asr kaldi-server speech-recognition speech-to-text
Last synced: 25 days ago
JSON representation
Server framework for Kaldi ASR Toolkit
- Host: GitHub
- URL: https://github.com/skit-ai/kaldi-serve
- Owner: skit-ai
- License: apache-2.0
- Created: 2019-10-17T14:16:27.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2023-09-17T21:09:39.000Z (over 1 year ago)
- Last Synced: 2025-03-25T09:18:16.419Z (about 1 month ago)
- Topics: asr, grpc, grpc-server, kaldi, kaldi-asr, kaldi-server, speech-recognition, speech-to-text
- Language: C++
- Homepage:
- Size: 18.7 MB
- Stars: 98
- Watchers: 20
- Forks: 24
- Open Issues: 17
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Kaldi-Serve
 
A plug-and-play abstraction over [Kaldi](https://kaldi-asr.org/) ASR toolkit, designed for ease of deployment and optimal runtime performance.
**Key Features**:
- Real-time streaming (uni & bi-directional) audio recognition.
- Thread-safe concurrent Decoder queue for server environments.
- RNNLM lattice rescoring.
- N-best alternatives with AM/LM costs, word-level timings and confidence scores.
- Easy extensibility for custom applications.
## Installation
### Dependencies
Make sure you have the following dependencies installed on your system before beginning the build process:
* g++ compiler (>=4.7) that supports C++11 std
* [CMake](https://cmake.org/install/) (>=3.13)
* [Kaldi](https://kaldi-asr.org/)
* [Boost C++](https://www.boost.org/) libraries
### Build from Source
Let's build the shared library:
```bash
cd build/
cmake ..
make -j${nproc}
```
You will find the the built shared library in `build/src/` to use for linking against custom applications.
#### Python bindings
We also provide python bindings for the library. You can find the build instructions [here](./python).
### Docker Image
#### Using pre-built images
You can also pull a pre-built docker image from our [Docker Hub repository](https://hub.docker.com/repository/docker/vernacularai/kaldi-serve):
```bash
docker pull vernacularai/kaldi-serve:latest
docker run -it -v /path/to/my/app:/home/app vernacularai/kaldi-serve:latest
```
You will find our headers in `/usr/include/kaldiserve` and the shared library `libkaldiserve.so` in `/usr/local/lib`.
#### Building the image
You can build the docker image using the [Dockerfile](./Dockerfile) provided.
```bash
docker build -t kaldi-serve:lib .
```
## Getting Started
### Usage
You can include the [headers](./include) and link the shared library you get after the build process, against your application and start using it.
### Plugins
It's also worth noting that there are a few [plugins](./plugins) we actively maintain and will keep adding to, that use the library:
- [gRPC Server](./plugins/grpc)
## License
This project is licensed under the Apache License version 2.0. Please see [LICENSE](./LICENSE) for more details.