mirror of
https://github.com/csukuangfj/kaldifeat.git
synced 2025-08-09 18:12:17 +00:00
kaldifeat
Wrap kaldi's feature computations to Python with PyTorch support.
Installation
From PyPi with pip
If you install kaldifeat
using pip
, it will also install
PyTorch 1.8.1. If this is not what you want, please install kaldifeat
from source (see below).
pip install kaldifeat
From source
The following are the commands to compile kaldifeat
from source.
We assume that you have installed cmake
and PyTorch.
cmake 3.11 is known to work. Other cmake versions may also work.
PyTorch 1.8.1 is known to work. Other PyTorch versions may also work.
mkdir /some/path
git clone https://github.com/csukuangfj/kaldifeat.git
cd kaldifeat
python setup.py install
To test whether kaldifeat
was installed successfully, you can run:
python3 -c "import kaldifeat; print(kaldifeat.__version__)"
Usage
Please refer to https://kaldifeat.readthedocs.io/en/latest/usage.html
for how to use kaldifeat
.
Description
Kaldi-compatible online & offline feature extraction with PyTorch, supporting CUDA, batch processing, chunk processing, and autograd - Provide C++ & Python API
cppfbankfeatures-extractionkaldimfcconline-feature-extractorplppythonpytorchstreaming-feature-extractor
Readme
Apache-2.0
11 MiB
Languages
C++
78%
Python
11.2%
CMake
6.7%
Shell
4.1%