open-speech / Kaldi Io
Licence: mit
c++ Kaldi IO lib (static and dynamic).
Stars: ✭ 22
Programming Languages
c
50402 projects - #5 most used programming language
Labels
Projects that are alternatives of or similar to Kaldi Io
opensnips
Open source projects related to Snips https://snips.ai/.
Stars: ✭ 50 (+127.27%)
Mutual labels: kaldi
Vosk Android Demo
Offline speech recognition for Android with Vosk library.
Stars: ✭ 271 (+1131.82%)
Mutual labels: kaldi
Awesome Kaldi
This is a list of features, scripts, blogs and resources for better using Kaldi ( http://kaldi-asr.org/ )
Stars: ✭ 393 (+1686.36%)
Mutual labels: kaldi
kaldi-alligner
scripts to align a given wave to its transcription using trained models by Kaldi
Stars: ✭ 24 (+9.09%)
Mutual labels: kaldi
Speech Aligner
speech-aligner,是一个从“人声语音”及其“语言文本”,产生音素级别时间对齐标注的工具。speech-aligner, is a tool that generate phoneme-level alignment between human speech and its transcription
Stars: ✭ 259 (+1077.27%)
Mutual labels: kaldi
React Transcript Editor
A React component to make correcting automated transcriptions of audio and video easier and faster. By BBC News Labs. - Work in progress
Stars: ✭ 285 (+1195.45%)
Mutual labels: kaldi
kaldi helpers
🙊 A set of scripts to use in preparing a corpus for speech-to-text processing with the Kaldi Automatic Speech Recognition Library.
Stars: ✭ 13 (-40.91%)
Mutual labels: kaldi
Docker Kaldi Gstreamer Server
Dockerfile for kaldi-gstreamer-server.
Stars: ✭ 266 (+1109.09%)
Mutual labels: kaldi
Zamia Speech
Open tools and data for cloudless automatic speech recognition
Stars: ✭ 374 (+1600%)
Mutual labels: kaldi
kaldi-timit-sre-ivector
Develop speaker recognition model based on i-vector using TIMIT database
Stars: ✭ 17 (-22.73%)
Mutual labels: kaldi
dropclass speaker
DropClass and DropAdapt - repository for the paper accepted to Speaker Odyssey 2020
Stars: ✭ 20 (-9.09%)
Mutual labels: kaldi
vosk-model-ru-adaptation
No description or website provided.
Stars: ✭ 19 (-13.64%)
Mutual labels: kaldi
Montreal Forced Aligner
Command line utility for forced alignment using Kaldi
Stars: ✭ 490 (+2127.27%)
Mutual labels: kaldi
KaldiBasedSpeakerVerification
Kaldi based speaker verification
Stars: ✭ 43 (+95.45%)
Mutual labels: kaldi
Vosk Server
WebSocket, gRPC and WebRTC speech recognition server based on Vosk and Kaldi libraries
Stars: ✭ 277 (+1159.09%)
Mutual labels: kaldi
Espresso
Espresso: A Fast End-to-End Neural Speech Recognition Toolkit
Stars: ✭ 808 (+3572.73%)
Mutual labels: kaldi
KALDI-IO lib
Kaldi c++ IO lib (static and dynamic).
It can be used as "kaldi I/O C++ API". Using this library, ".ark,.scp" formats can be read and write, and kaldi::[Matrix|Vector|..] is available.
Compile
- requirements:
- cmake >= 3.0
- one of blas math lib:
- mkl (recommended)
- install conda, and use it to install mkl:
conda install mkl
(mkl is installed with conda by default) - when cmake,
conda
should be in your path
- install conda, and use it to install mkl:
- atlas
- ubuntu:
sudo apt-get install libatlas3-base
- when cmake, it maynot find your atlas automatically, thus you need set the math lib path as in "#About" below.
- ubuntu:
- Accelerate framework (need do nothing for "macOS/Darwin")
- ...
- mkl (recommended)
cd kaldi-io
mkdir build && cd build
cmake -DCMAKE_INSTALL_PREFIX=.. .. # set install prefix as '../kaldi-io'
make
make install
Results:
- kaldi-io/lib:
- libkaldi_io_static.a
- libkaldi_io_shared.so
- kaldi-io/include:
- kaldi-io.h (the one big header, can be modified by need)
- sub header dirs...
Usage
- Include the header dir and link the lib
About
-
The math dependency is solved automatically from system path, by cmake
find_package
(cmake/Modules/FindBLAS.cmake)- MKL/ATLAS/Accelerate(osx) is prefered
- if you want to set specific math lib:
# make sure build dir is clean cmake -DBLAS_VENDORS=[ATLAS|MKL|OPEN|..] .. # additioinally, custom math lib search path can be set, for example: cmake -DBLAS_VENDORS=ATLAS -DBLAS_ATLAS_LIB_DIRS=.../atlas/build/lib .. cmake -DBLAS_VENDORS=MKL -DBLAS_MKL_LIB_DIRS=/opt/intel/mkl/lib/intel64 ..
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].