PyKaldi is a Python scripting layer for the Kaldi speech recognition toolkit. It provides easy-to-use, low-overhead, first-class Python wrappers for the C++ code in Kaldi and OpenFst libraries. You can use PyKaldi to write Python code for things that would otherwise require writing C++ code such as calling low-level Kaldi functions, manipulating Kaldi and OpenFst objects in code or implementing new Kaldi tools.
You can think of Kaldi as a large box of legos that you can mix and match to build custom speech recognition solutions. The best way to think of PyKaldi is as a supplement, a sidekick if you will, to Kaldi. In fact, PyKaldi is at its best when it is used alongside Kaldi. To that end, replicating the functionality of myriad command-line tools, utility scripts and shell-level recipes provided by Kaldi is a non-goal for the PyKaldi project.
Like Kaldi, PyKaldi is primarily intended for speech recognition researchers and professionals. It is jam packed with goodies that one would need to build Python software taking advantage of the vast collection of utilities, algorithms and data structures provided by Kaldi and OpenFst libraries.
If you are not familiar with FST-based speech recognition or have no interest in
having access to the guts of Kaldi and OpenFst in Python, but only want to run a
pre-trained Kaldi system as part of your Python application, do not fret.
PyKaldi includes a number of high-level application oriented modules, such as
asr, alignment and segmentation, that should be accessible to most
Python programmers.
If you are interested in using PyKaldi for research or building advanced ASR
applications, you are in luck. PyKaldi comes with everything you need to read,
write, inspect, manipulate or visualize Kaldi and OpenFst objects in Python. It
includes Python wrappers for most functions and methods that are part of the
public APIs of Kaldi and OpenFst C++ libraries. If you want to read/write files
that are produced/consumed by Kaldi tools, check out I/O and table utilities in
the util package. If you want to work with Kaldi matrices and vectors, e.g.
convert them to NumPy ndarrays and vice versa, check out the matrix
package. If you want to use Kaldi for feature extraction and transformation,
check out the feat, ivector and transform packages. If you want to
work with lattices or other FST structures produced/consumed by Kaldi tools,
check out the fstext, lat and kws packages. If you want low-level
access to Gaussian mixture models, hidden Markov models or phonetic decision
trees in Kaldi, check out the gmm, sgmm2, hmm, and tree
packages. If you want low-level access to Kaldi neural network models, check out
the nnet3, cudamatrix and chain packages. If you want to use the
decoders and language modeling utilities in Kaldi, check out the decoder,
lm, rnnlm, tfrnnlm and online2 packages.
Interested readers who would like to learn more about Kaldi and PyKaldi might find the following resources useful:
- Kaldi Docs: Read these to learn more about Kaldi.
- PyKaldi Docs: Consult these to learn more about the PyKaldi API.
- PyKaldi Examples: Check these out to see PyKaldi in action.
- PyKaldi Paper: Read this to learn more about the design of PyKaldi.
Since automatic speech recognition (ASR) in Python is undoubtedly the "killer app" for PyKaldi, we will go over a few ASR scenarios to get a feel for the PyKaldi API. We should note that PyKaldi does not provide any high-level utilities for training ASR models, so you need to train your models using Kaldi recipes or use pre-trained models available online. The reason why this is so is simply because there is no high-level ASR training API in Kaldi C++ libraries. Kaldi ASR models are trained using complex shell-level recipes that handle everything from data preparation to the orchestration of myriad Kaldi executables used in training. This is by design and unlikely to change in the future. PyKaldi does provide wrappers for the low-level ASR training utilities in Kaldi C++ libraries but those are not really useful unless you want to build an ASR training pipeline in Python from basic building blocks, which is no easy task. Continuing with the lego analogy, this task is akin to building this given access to a truck full of legos you might need. If you are crazy enough to try though, please don't let this paragraph discourage you. Before we started building PyKaldi, we thought that was a mad man's task too.
PyKaldi asr module includes a number of easy-to-use, high-level classes to
make it dead simple to put together ASR systems in Python. Ignoring the
boilerplate code needed for setting things up, doing ASR with PyKaldi can be as
simple as the following snippet of code:
asr = SomeRecognizer.from_files("final.mdl", "HCLG.fst", "words.txt", opts)
with SequentialMatrixReader("ark:feats.ark") as feats_reader:
for key, feats in feats_reader:
out = asr.decode(feats)
print(key, out["text"])In this simplified example, we first instantiate a hypothetical recognizer
SomeRecognizer with the paths for the model final.mdl, the decoding graph
HCLG.fst and the symbol table words.txt. The opts object contains the
configuration options for the recognizer. Then, we instantiate a PyKaldi table
reader SequentialMatrixReader for reading the feature
matrices stored in the Kaldi archive feats.ark. Finally,
we iterate over the feature matrices and decode them one by one. Here we are
simply printing the best ASR hypothesis for each utterance so we are only
interested in the "text" entry of the output dictionary out. Keep in mind
that the output dictionary contains a bunch of other useful entries, such as the
frame level alignment of the best hypothesis and a weighted lattice representing
the most likely hypotheses. Admittedly, not all ASR pipelines will be as simple
as this example, but they will often have the same overall structure. In the
following sections, we will see how we can adapt the code given above to
implement more complicated ASR pipelines.
This is the most common scenario. We want to do offline ASR using pre-trained Kaldi models, such as ASpIRE chain models. Here we are using the term "models" loosely to refer to everything one would need to put together an ASR system. In this specific example, we are going to need:
- a neural network acoustic model,
- a transition model,
- a decoding graph,
- a word symbol table,
- and a couple of feature extraction configs.
Note that you can use this example code to decode with ASpIRE chain models.
from kaldi.asr import NnetLatticeFasterRecognizer
from kaldi.decoder import LatticeFasterDecoderOptions
from kaldi.nnet3 import NnetSimpleComputationOptions
from kaldi.util.table import SequentialMatrixReader, CompactLatticeWriter
# Set the paths and read/write specifiers
model_path = "models/aspire/final.mdl"
graph_path = "models/aspire/graph_pp/HCLG.fst"
symbols_path = "models/aspire/graph_pp/words.txt"
feats_rspec = ("ark:compute-mfcc-feats --config=models/aspire/conf/mfcc.conf "
"scp:wav.scp ark:- |")
ivectors_rspec = (feats_rspec + "ivector-extract-online2 "
"--config=models/aspire/conf/ivector_extractor.conf "
"ark:spk2utt ark:- ark:- |")
lat_wspec = "ark:| gzip -c > lat.gz"
# Instantiate the recognizer
decoder_opts = LatticeFasterDecoderOptions()
decoder_opts.beam = 13
decoder_opts.max_active = 7000
decodable_opts = NnetSimpleComputationOptions()
decodable_opts.acoustic_scale = 1.0
decodable_opts.frame_subsampling_factor = 3
asr = NnetLatticeFasterRecognizer.from_files(
model_path, graph_path, symbols_path,
decoder_opts=decoder_opts, decodable_opts=decodable_opts)
# Extract the features, decode and write output lattices
with SequentialMatrixReader(feats_rspec) as feats_reader, \
SequentialMatrixReader(ivectors_rspec) as ivectors_reader, \
CompactLatticeWriter(lat_wspec) as lat_writer:
for (fkey, feats), (ikey, ivectors) in zip(feats_reader, ivectors_reader):
assert(fkey == ikey)
out = asr.decode((feats, ivectors))
print(fkey, out["text"])
lat_writer[fkey] = out["lattice"]The fundamental difference between this example and the short snippet from last
section is that for each utterance we are reading the raw audio data from disk
and computing two feature matrices on the fly instead of reading a single
precomputed feature matrix from disk. The script file
wav.scp contains a list of WAV files corresponding to the utterances we want
to decode. The additional feature matrix we are extracting contains online
i-vectors that are used by the neural network acoustic model to perform channel
and speaker adaptation. The speaker-to-utterance map
spk2utt is used for accumulating separate statistics for each speaker in
online i-vector extraction. It can be a simple identity mapping if the speaker
information is not available. We pack the MFCC features and the i-vectors into a
tuple and pass this tuple to the recognizer for decoding. The neural network
recognizers in PyKaldi know how to handle the additional i-vector features when
they are available. The model file final.mdl contains both the transition
model and the neural network acoustic model. The NnetLatticeFasterRecognizer
processes feature matrices by first computing phone log-likelihoods using the
neural network acoustic model, then mapping those to transition log-likelihoods
using the transition model and finally decoding transition log-likelihoods into
word sequences using the decoding graph HCLG.fst, which has transition
IDs on its input labels and word IDs on its output labels. After decoding, we save the lattice
generated by the recognizer to a Kaldi archive for future processing.
This example also illustrates the powerful I/O mechanisms
provided by Kaldi. Instead of implementing the feature extraction pipelines in
code, we define them as Kaldi read specifiers and compute the feature matrices
simply by instantiating PyKaldi table readers and
iterating over them. This is not only the simplest but also the fastest way of
computing features with PyKaldi since the feature extraction pipeline is run in
parallel by the operating system. Similarly, we use a Kaldi write specifier to
instantiate a PyKaldi table writer which writes output
lattices to a compressed Kaldi archive. Note that for these to work, we need
compute-mfcc-feats, ivector-extract-online2 and gzip to be on our PATH.
This is similar to the previous scenario, but instead of a Kaldi acoustic model, we use a PyTorch acoustic model. After computing the features as before, we convert them to a PyTorch tensor, do the forward pass using a PyTorch neural network module outputting phone log-likelihoods and finally convert those log-likelihoods back into a PyKaldi matrix for decoding. The recognizer uses the transition model to automatically map phone IDs to transition IDs, the input labels on a typical Kaldi decoding graph.
from kaldi.asr import MappedLatticeFasterRecognizer
from kaldi.decoder import LatticeFasterDecoderOptions
from kaldi.matrix import Matrix
from kaldi.util.table import SequentialMatrixReader, CompactLatticeWriter
from models import AcousticModel # Import your PyTorch model
import torch
# Set the paths and read/write specifiers
acoustic_model_path = "models/aspire/model.pt"
transition_model_path = "models/aspire/final.mdl"
graph_path = "models/aspire/graph_pp/HCLG.fst"
symbols_path = "models/aspire/graph_pp/words.txt"
feats_rspec = ("ark:compute-mfcc-feats --config=models/aspire/conf/mfcc.conf "
"scp:wav.scp ark:- |")
lat_wspec = "ark:| gzip -c > lat.gz"
# Instantiate the recognizer
decoder_opts = LatticeFasterDecoderOptions()
decoder_opts.beam = 13
decoder_opts.max_active = 7000
asr = MappedLatticeFasterRecognizer.from_files(
transition_model_path, graph_path, symbols_path, decoder_opts=decoder_opts)
# Instantiate the PyTorch acoustic model (subclass of torch.nn.Module)
model = AcousticModel(...)
model.load_state_dict(torch.load(acoustic_model_path))
model.eval()
# Extract the features, decode and write output lattices
with SequentialMatrixReader(feats_rspec) as feats_reader, \
CompactLatticeWriter(lat_wspec) as lat_writer:
for key, feats in feats_reader:
feats = torch.from_numpy(feats.numpy()) # Convert to PyTorch tensor
loglikes = model(feats) # Compute log-likelihoods
loglikes = Matrix(loglikes.numpy()) # Convert to PyKaldi matrix
out = asr.decode(loglikes)
print(key, out["text"])
lat_writer[key] = out["lattice"]This section is a placeholder. Check out this script in the meantime.
Lattice rescoring is a standard technique for using large n-gram language models or recurrent neural network language models (RNNLMs) in ASR. In this example, we rescore lattices using a Kaldi RNNLM. We first instantiate a rescorer by providing the paths for the models. Then we use a table reader to iterate over the lattices we want to rescore and finally we use a table writer to write rescored lattices back to disk.
from kaldi.asr import LatticeRnnlmPrunedRescorer
from kaldi.fstext import SymbolTable
from kaldi.rnnlm import RnnlmComputeStateComputationOptions
from kaldi.util.table import SequentialCompactLatticeReader, CompactLatticeWriter
# Set the paths, extended filenames and read/write specifiers
symbols_path = "models/tedlium/config/words.txt"
old_lm_path = "models/tedlium/data/lang_nosp/G.fst"
word_feats_path = "models/tedlium/word_feats.txt"
feat_embedding_path = "models/tedlium/feat_embedding.final.mat"
word_embedding_rxfilename = ("rnnlm-get-word-embedding %s %s - |"
% (word_feats_path, feat_embedding_path))
rnnlm_path = "models/tedlium/final.raw"
lat_rspec = "ark:gunzip -c lat.gz |"
lat_wspec = "ark:| gzip -c > rescored_lat.gz"
# Instantiate the rescorer
symbols = SymbolTable.read_text(symbols_path)
opts = RnnlmComputeStateComputationOptions()
opts.bos_index = symbols.find_index("<s>")
opts.eos_index = symbols.find_index("</s>")
opts.brk_index = symbols.find_index("<brk>")
rescorer = LatticeRnnlmPrunedRescorer.from_files(
old_lm_path, word_embedding_rxfilename, rnnlm_path, opts=opts)
# Read the lattices, rescore and write output lattices
with SequentialCompactLatticeReader(lat_rspec) as lat_reader, \
CompactLatticeWriter(lat_wspec) as lat_writer:
for key, lat in lat_reader:
lat_writer[key] = rescorer.rescore(lat)Notice the extended filename we used to compute the word embeddings from the
word features and the feature embeddings on the fly. Also of note are the
read/write specifiers we used to transparently decompress/compress the lattice
archives. For these to work, we need rnnlm-get-word-embedding, gunzip and
gzip to be on our PATH.
PyKaldi aims to bridge the gap between Kaldi and all the nice things Python has to offer. It is more than a collection of bindings into Kaldi libraries. It is a scripting layer providing first class support for essential Kaldi and OpenFst types in Python. PyKaldi vector and matrix types are tightly integrated with NumPy. They can be seamlessly converted to NumPy arrays and vice versa without copying the underlying memory buffers. PyKaldi FST types, including Kaldi style lattices, are first class citizens in Python. The API for the user facing FST types and operations is almost entirely defined in Python mimicking the API exposed by pywrapfst, the official Python wrapper for OpenFst.
PyKaldi harnesses the power of CLIF to wrap Kaldi and OpenFst C++ libraries using simple API descriptions. The CPython extension modules generated by CLIF can be imported in Python to interact with Kaldi and OpenFst. While CLIF is great for exposing existing C++ API in Python, the wrappers do not always expose a "Pythonic" API that is easy to use from Python. PyKaldi addresses this by extending the raw CLIF wrappers in Python (and sometimes in C++) to provide a more "Pythonic" API. Below figure illustrates where PyKaldi fits in the Kaldi ecosystem.
PyKaldi has a modular design which makes it easy to maintain and extend. Source files are organized in a directory tree that is a replica of the Kaldi source tree. Each directory defines a subpackage and contains only the wrapper code written for the associated Kaldi library. The wrapper code consists of:
-
CLIF C++ API descriptions defining the types and functions to be wrapped and their Python API,
-
C++ headers defining the shims for Kaldi code that is not compliant with the Google C++ style expected by CLIF,
-
Python modules grouping together related extension modules generated with CLIF and extending the raw CLIF wrappers to provide a more "Pythonic" API.
You can read more about the design and technical details of PyKaldi in our paper.
The following table shows the status of each PyKaldi package (we currently do not plan to add support for nnet, nnet2 and online) along the following dimensions:
- Wrapped?: If there are enough CLIF files to make the package usable in Python.
- Pythonic?: If the package API has a "Pythonic" look-and-feel.
- Documentation?: If there is documentation beyond what is automatically generated by CLIF. Single checkmark indicates that there is not much additional documentation (if any). Three checkmarks indicates that package documentation is complete (or near complete).
- Tests?: If there are any tests for the package.
| Package | Wrapped? | Pythonic? | Documentation? | Tests? |
|---|---|---|---|---|
| base | ✔ | ✔ | ✔ ✔ ✔ | ✔ |
| chain | ✔ | ✔ | ✔ ✔ ✔ | |
| cudamatrix | ✔ | ✔ | ✔ | |
| decoder | ✔ | ✔ | ✔ ✔ ✔ | |
| feat | ✔ | ✔ | ✔ ✔ ✔ | |
| fstext | ✔ | ✔ | ✔ ✔ ✔ | |
| gmm | ✔ | ✔ | ✔ ✔ | ✔ |
| hmm | ✔ | ✔ | ✔ ✔ ✔ | ✔ |
| ivector | ✔ | ✔ | ||
| kws | ✔ | ✔ | ✔ ✔ ✔ | |
| lat | ✔ | ✔ | ✔ ✔ ✔ | |
| lm | ✔ | ✔ | ✔ ✔ ✔ | |
| matrix | ✔ | ✔ | ✔ ✔ ✔ | ✔ |
| nnet3 | ✔ | ✔ | ✔ | |
| online2 | ✔ | ✔ | ✔ ✔ ✔ | |
| rnnlm | ✔ | ✔ | ✔ ✔ ✔ | |
| sgmm2 | ✔ | ✔ | ||
| tfrnnlm | ✔ | ✔ | ✔ ✔ ✔ | |
| transform | ✔ | ✔ | ✔ | |
| tree | ✔ | ✔ | ||
| util | ✔ | ✔ | ✔ ✔ ✔ | ✔ |
If you are using a relatively recent Linux or macOS, such as Ubuntu >= 16.04, CentOS >= 7 or macOS >= 10.13, you should be able to install PyKaldi without too much trouble. Otherwise, you will likely need to tweak the installation scripts.
You can now download official whl packages from our github release page. We have whl packages for Python 3.7, 3.8, ... , 3.11 on Linux and a few (experimental) builds for Mac M1/M2.
If you decide to use a whl package then you can skip the next sections and head straight to "Starting a new project with a pykaldi whl package" to setup your project. Note that you still need to compile a PyKaldi-compatible version of Kaldi.
To install and build PyKaldi from source, follow the steps given below.
git clone https://github.com/pykaldi/pykaldi.git
cd pykaldiAlthough it is not required, we recommend installing PyKaldi and all of its Python dependencies inside a new isolated Python environment. If you do not want to create a new Python environment, you can skip the rest of this step.
You can use any tool you like for creating a new Python environment. Here we
use virtualenv, but you can use another tool like conda if you prefer that.
Make sure you activate the new Python environment before continuing with the
rest of the installation.
virtualenv env
source env/bin/activateRunning the commands below will install the system packages needed for building PyKaldi from source.
# Ubuntu
sudo apt-get install autoconf automake cmake curl g++ git graphviz \
libatlas3-base libtool make pkg-config subversion unzip wget zlib1g-dev
# macOS
brew install automake cmake git graphviz libtool pkg-config wget gnu-sed openblas subversion
PATH="/opt/homebrew/opt/gnu-sed/libexec/gnubin:$PATH"Running the commands below will install the Python packages needed for building PyKaldi from source.
pip install --upgrade pip
pip install --upgrade setuptools
pip install numpy pyparsing
pip install ninja # not required but strongly recommendedIn addition to above listed packages, we also need PyKaldi compatible installations of the following software:
-
Google Protobuf, recommended v3.5.0. Both the C++ library and the Python package must be installed.
-
PyKaldi compatible fork of CLIF. To streamline PyKaldi development, we made some changes to CLIF codebase. We are hoping to upstream these changes over time. These changes are in the pykaldi branch:
# This command will be automatically run for you in the tools install scripts.
git clone -b pykaldi https://github.com/pykaldi/clif- PyKaldi compatible fork of Kaldi. To comply with CLIF requirements we had to make some changes to Kaldi codebase. We are hoping to upstream these changes over time.These changes are in the pykaldi branch:
# This command will be automatically run for you in the tools install scripts.
git clone -b pykaldi https://github.com/pykaldi/kaldiYou can use the scripts in the tools directory to install or update these
software locally. Make sure you check the output of these scripts. If you do not
see Done installing {protobuf,CLIF,Kaldi} printed at the very end, it means
that installation has failed for some reason.
cd tools
./check_dependencies.sh # checks if system dependencies are installed
./install_protobuf.sh # installs both the C++ library and the Python package
./install_clif.sh # installs both the C++ library and the Python package
./install_kaldi.sh # installs the C++ library
cd ..Note, if you are compiling Kaldi on Apple Silicion and ./install_kaldi.sh gets stuck right at the beginning compiling sctk, you might need to remove -march=native from tools/kaldi/tools/Makefile, e.g. by uncommeting it in this line like this:
SCTK_CXFLAGS = -w #-march=nativeIf Kaldi is installed inside the tools directory and all Python dependencies
(numpy, pyparsing, pyclif, protobuf) are installed in the active Python
environment, you can install PyKaldi with the following command.
python setup.py installOnce installed, you can run PyKaldi tests with the following command.
python setup.py testYou can then also create a whl package. The whl package makes it easy to install pykaldi into a new project environment for your speech project.
python setup.py bdist_wheelThe whl file can then be found in the "dist" folder. The whl filename depends on the pykaldi version, your Python version and your architecture. For a Python 3.9 build on x86_64 with pykaldi 0.2.2 it may look like: dist/pykaldi-0.2.2-cp39-cp39-linux_x86_64.whl
Create a new project folder, for example:
mkdir -p ~/projects/myASR
cd ~/projects/myASRCreate and activate a virtual environment with the same Python version as the whl package, e.g for Python 2.9:
virtualenv -p /usr/bin/python3.9 myasr_env
. myasr_env/bin/activateInstall numpy and pykaldi into your myASR environment:
pip3 install numpy
pip3 install pykaldi-0.2.2-cp39-cp39-linux_x86_64.whl Copy pykaldi/tools/install_kaldi.sh to your myASR project. Use the install_kaldi.sh script to install a pykaldi compatible kaldi version for your project:
./install_kaldi.shCopy pykaldi/tools/path.sh to your project. Path.sh is used to make pykaldi find the Kaldi libraries and binaries in the kaldi folder. Source path.sh with:
. path.shCongratulations, you are ready to use pykaldi in your project!
Note: Anytime you open a new shell, you need to source the project environment and path.sh:
. myasr_env/bin/activate
. path.shNote: Unfortunatly, the PyKaldi Conda packages are outdated. If you would like to maintain it, please get in touch with us.
To install PyKaldi with CUDA support:
conda install -c pykaldi pykaldiTo install PyKaldi without CUDA support (CPU only):
conda install -c pykaldi pykaldi-cpuNote that PyKaldi conda package does not provide Kaldi executables. If you would like to use Kaldi executables along with PyKaldi, e.g. as part of read/write specifiers, you need to install Kaldi separately.
Note: The docker instructions below may be outdated. If you would like to maintain a docker image for PyKaldi, please get in touch with us.
If you would like to use PyKaldi inside a Docker container, follow the
instructions in the docker folder.
By default, PyKaldi install command uses all available (logical) processors to accelerate the build process. If the size of the system memory is relatively small compared to the number of processors, the parallel compilation/linking jobs might end up exhausting the system memory and result in swapping. You can limit the number of parallel jobs used for building PyKaldi as follows:
MAKE_NUM_JOBS=2 python setup.py installWe have no idea what is needed to build PyKaldi on Windows. It would probably require lots of changes to the build system.
At the moment, PyKaldi is not compatible with the upstream Kaldi repository. You need to build it against our Kaldi fork.
If you already have a compatible Kaldi installation on your system, you do not
need to install a new one inside the pykaldi/tools directory. Instead, you
can simply set the following environment variable before running the PyKaldi
installation command.
export KALDI_DIR=<directory where Kaldi is installed, e.g. "$HOME/tools/kaldi">At the moment, PyKaldi is not compatible with the upstream CLIF repository. You need to build it using our CLIF fork.
If you already have a compatible CLIF installation on your system, you do not
need to install a new one inside the pykaldi/tools directory. Instead, you
can simply set the following environment variables before running the PyKaldi
installation command.
export PYCLIF=<path to pyclif executable, e.g. "$HOME/anaconda3/envs/clif/bin/pyclif">
export CLIF_MATCHER=<path to clif-matcher executable, e.g. "$HOME/anaconda3/envs/clif/clang/bin/clif-matcher">While the need for updating Protobuf and CLIF should not come up very often, you
might want or need to update Kaldi installation used for building PyKaldi.
Rerunning the relevant install script in tools directory should update the
existing installation. If this does not work, please open an issue.
PyKaldi tfrnnlm package is built automatically along with the rest of PyKaldi
if kaldi-tensorflow-rnnlm library can be found among Kaldi libraries. After
building Kaldi, go to KALDI_DIR/src/tfrnnlm/ directory and follow the
instructions given in the Makefile. Make sure the symbolic link for the
kaldi-tensorflow-rnnlm library is added to the KALDI_DIR/src/lib/ directory.
Shennong - a toolbox for speech features extraction, like MFCC, PLP etc. using PyKaldi.
Kaldi model server - a threaded kaldi model server for live decoding. Can directly decode speech from your microphone with a nnet3 compatible model. Example models for English and German are available. Uses the PyKaldi online2 decoder.
MeetingBot - example of a web application for meeting transcription and summarization that makes use of a pykaldi/kaldi-model-server backend to display ASR output in the browser.
Subtitle2go - automatic subtitle generation for any media file. Uses PyKaldi for ASR with a batch decoder.
If you have a cool open source project that makes use of PyKaldi that you'd like to showcase here, let us know!
If you use PyKaldi for research, please cite our paper as follows:
@inproceedings{pykaldi,
title = {PyKaldi: A Python Wrapper for Kaldi},
author = {Dogan Can and Victor R. Martinez and Pavlos Papadopoulos and
Shrikanth S. Narayanan},
booktitle={Acoustics, Speech and Signal Processing (ICASSP),
2018 IEEE International Conference on},
year = {2018},
organization = {IEEE}
}
We appreciate all contributions! If you find a bug, feel free to open an issue or a pull request. If you would like to request or add a new feature please open an issue for discussion.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent