All Projects → loretoparisi → htk

loretoparisi / htk

Licence: other
HTK Toolkit with Linux 64 bit and Docker support

Programming Languages

c
50402 projects - #5 most used programming language
TeX
3793 projects

Projects that are alternatives of or similar to htk

voce-browser
Voice Controlled Chromium Web Browser
Stars: ✭ 34 (+142.86%)
Mutual labels:  speech-recognition, speech-to-text
SpeechToText
Speech To Text in Android
Stars: ✭ 53 (+278.57%)
Mutual labels:  speech-recognition, speech-to-text
spokestack-ios
Spokestack: give your iOS app a voice interface!
Stars: ✭ 27 (+92.86%)
Mutual labels:  speech-recognition, speech-to-text
Unity live caption
Use Google Speech-to-Text API to do real-time live stream caption on Unity! Best when combined with your virtual character!
Stars: ✭ 26 (+85.71%)
Mutual labels:  speech-recognition, speech-to-text
speech-to-text-code-pattern
React app using the Watson Speech to Text service to transform voice audio into written text.
Stars: ✭ 37 (+164.29%)
Mutual labels:  speech-recognition, speech-to-text
kaldi-long-audio-alignment
Long audio alignment using Kaldi
Stars: ✭ 21 (+50%)
Mutual labels:  speech-recognition, speech-to-text
deepspeech
A PyTorch implementation of DeepSpeech and DeepSpeech2.
Stars: ✭ 45 (+221.43%)
Mutual labels:  speech-recognition, speech-to-text
kaldi ag training
Docker image and scripts for training finetuned or completely personal Kaldi speech models. Particularly for use with kaldi-active-grammar.
Stars: ✭ 14 (+0%)
Mutual labels:  speech-recognition, speech-to-text
Deep-learning-And-Paper
【仅作为交流学习使用】机器智能--相关书目及经典论文包括AutoML、情感分类、语音识别、声纹识别、语音合成实验代码等
Stars: ✭ 62 (+342.86%)
Mutual labels:  speech-recognition, speech-to-text
revai-node-sdk
Node.js SDK for the Rev AI API
Stars: ✭ 21 (+50%)
Mutual labels:  speech-recognition, speech-to-text
open-speech-corpora
💎 A list of accessible speech corpora for ASR, TTS, and other Speech Technologies
Stars: ✭ 841 (+5907.14%)
Mutual labels:  speech-recognition, speech-to-text
speech to text
how to use the Google Cloud Speech API to transcribe audio/video files.
Stars: ✭ 35 (+150%)
Mutual labels:  speech-recognition, speech-to-text
Chinese-automatic-speech-recognition
Chinese speech recognition
Stars: ✭ 147 (+950%)
Mutual labels:  speech-recognition, speech-to-text
speech-to-text
mixlingual speech recognition system; hybrid (GMM+NNet) model; Kaldi + Keras
Stars: ✭ 61 (+335.71%)
Mutual labels:  speech-recognition, speech-to-text
scripty
Speech to text bot for Discord using Mozilla's DeepSpeech
Stars: ✭ 14 (+0%)
Mutual labels:  speech-recognition, speech-to-text
deep avsr
A PyTorch implementation of the Deep Audio-Visual Speech Recognition paper.
Stars: ✭ 104 (+642.86%)
Mutual labels:  speech-recognition, speech-to-text
Inimesed
An Android app that lets you search your contacts by voice. Internet not required. Based on Pocketsphinx. Uses Estonian acoustic models.
Stars: ✭ 65 (+364.29%)
Mutual labels:  speech-recognition, speech-to-text
PCPM
Presenting Collection of Pretrained Models. Links to pretrained models in NLP and voice.
Stars: ✭ 21 (+50%)
Mutual labels:  speech-recognition, speech-to-text
deepspeech.mxnet
A MXNet implementation of Baidu's DeepSpeech architecture
Stars: ✭ 82 (+485.71%)
Mutual labels:  speech-recognition, speech-to-text
vosk-asterisk
Speech Recognition in Asterisk with Vosk Server
Stars: ✭ 52 (+271.43%)
Mutual labels:  speech-recognition, speech-to-text
	Hidden Markov Toolkit (HTK) 3.4.1

Use of this software is governed by a license agreement, the terms and conditions of which are set forth in the file LICENSE in the top-level HTK installation directory. Please read this file carefully as use of this software implies acceptance of the conditions described therein.

Introduction

HTK is a toolkit for use in research into automatic speech recognition and has been developed by the Machine Intelligence Laboratory (formerly know as the Speech Vision Robotics Group) at the Cambridge University Engineering Department (http://mi.eng.cam.ac.uk) and Entropic Ltd (http://www.entropic.com).

Please visit the HTK homepage at the following address for more information about HTK:

http://htk.eng.cam.ac.uk/

A number of mailing lists have been established to help users build and understand HTK, for details see

http://htk.eng.cam.ac.uk/mailing/subscribe_mail.shtml

License

HTK is made available free of charge and can be downloaded from the website mentioned above. However it may not be redistributed, i.e. you must register at the website and download it from there. Details about the terms under which HTK is made available can be found in the LICENSE file.

Docker

Why Docker? I have added docker support since the HTK source code is out of date and does not compile on modern x86_64 architectures due to minor issues. To build this docker image

git clone https://github.com/loretoparisi/htk.git \
cd htk/ \
docker build -t htk-docker . \
docker run --rm -it htk-docker

Compiling & Installing HTK under UNIX/Linux, OS X or Cygwin

After unpacking the sources, cd to the htk directory.

There are now two ways to install HTK, the "traditional" and the "new". Up to now HTK has always installed its tools as they were built, and installed them to a directory such as "bin.linux" so that binaries for different architectures can be installed in a home directory say. If you want to install in this way, please add the option "--enable-trad-htk" when you run configure.

The "new" method installs by default into /usr/local/bin (equivalent to a configure option of "--prefix=/usr/local").

  1. decide which of the above methods you wish to use
  2. cd to htk, then run ./configure (with appropriate options, run "./configure --help" if unsure). If you don't want to build the programs in HLMTools add the --disable-hlmtools option.
  3. make all
  4. make install

Running "make install" will install them. This step may need to be done as root, if you are not installing them in your home directory.

Notes for particular Unix variants: Solaris: if "make" isn't installed you may need to add /opt/sfw/bin and /usr/ccs/bin to your path and run "./configure MAKE=gmake" with any other options you require. Then run "gmake" instead of "make", alternatively you can create a symbolic link called "make" somewhere it your path to /opt/sfw/bin/gmake

HDecode

If you are also building HDecode (available from the HTK website, under a different licence from HTK), you will firstly need to unpack the HDecode source code (in the same directory in which you unpacked the HTK sources). Follow the steps above for building HTK first, then add these steps to the build process:

  1. make hdecode
  2. make install-hdecode

Compiling & Installing HTK under Windows

Prerequisites: * HTK has been verified to compile using Microsoft Visual Studio. * For testing, you will require a Perl interpreter such as ActivePerl.
* You will need a tool such as 7-zip or winzip (commercial) for unpacking the HTK source code archive. * It is helpful if you have some familiarity with using the DOS command line interface, as you will need to interact with it in order to compile, install and run HTK. * Ensure that your PATH contains C:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\bin Or if you are using older versions: C:\Program Files\Microsoft Visual Studio\VC98\bin

Compilation:

  1. Unpack the HTK sources using 7-zip.

  2. Open a DOS command window: Click Start, select Run type cmd at the prompt and click OK.

  3. cd into the directory in which you unpacked the sources.

  4. cd into the htk directory. Type:

    cd htk

  5. Create a directory for the library and tools. Type:

    mkdir bin.win32

  6. Run VCVARS32 (it should be in your path, see prerequisites above)

  7. Build the HTK Library, which provides the common functionality used by the HTK Tools. Enter the following commands:

    cd HTKLib nmake /f htk_htklib_nt.mkf all cd ..

  8. Build the HTK Tools

    cd HTKTools nmake /f htk_htktools_nt.mkf all cd .. cd HLMLib nmake /f htk_hlmlib_nt.mkf all cd .. cd HLMTools nmake /f htk_hlmtools_nt.mkf all cd ..

Installation: The HTK tools have now been built and are in the bin.win32 directory. You should add this directory to your PATH, so that you can run them easily from the command line in future.

Testing the Installation

Among the samples on the HTK website you'll find the HTKDemo package that can be used to test your installation. See http://htk.eng.cam.ac.uk/download.shtml for download instructions.

As an initial test of the installation please run the HTK demonstration using the configuration file HTKDemo/configs/monPlainM1S1.dcf. There is a README file in the HTKDemo directory explaining the operation of the demonstration in detail but, in short, you need to run the demonstration script passing it the configuration file configs/monPlainM1S1.dcf as input. To test the language modelling tools you should follow the tutorial in the HTK book, using the files in the LMTutorial/ directory.

Before running the demo make sure you have compiled all the HTK tools and the executables are in your PATH, i.e. just typing 'HInit' at the commandline prints a short usage summary. To run the demonstration type:

$ cd HTKDemo $ ./runDemo configs/monPlainM1S1.dcf

The recognition results obtained should match the following.

On the training set:

------------------------ Overall Results --------------------------
SENT: %Correct=0.00 [H=0, S=7, N=7]
WORD: %Corr=77.63, Acc=74.89 [H=170, D=37, S=12, I=6, N=219]
===================================================================

On the test set:

------------------------ Overall Results --------------------------
SENT: %Correct=0.00 [H=0, S=3, N=3]
WORD: %Corr=63.91, Acc=59.40 [H=85, D=35, S=13, I=6, N=133]
===================================================================

NB to run this demo under Windows you must have perl installed and you need to invoke perl explicitly. See http://www.perl.org/ to download the perl distribution. The script runDemo.pl should be used in place of runDemo i.e. to run the test above type

perl runDemo.pl configs\monPlainM1S1.dcf

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].