A GUI for Speect
DeSpeect is a graphical user interface for Speect.
Speect is a multilingual TTS(text-to-speech) system, initially developed by The Human Language Technologies group at the Meraka Institute, CSIR, South Africa (more information here).
DeSpeect requires the Speect version maintained and developed by MIVOQ(Github repository here).
The following dependencies must be fulfilled in order to successfully build Despeect:
- Qt 5.10.1
- g++ 5.4.0
- git 2.7.4
- curl 7.4.0
- swig 3.0.8
- libXML2 2.9.3
- Python 2.7.11
- autoreconf 5.9
- libGL 10.2.0
- CMake 2.8
DeSpeect has been developed and tested on Ubuntu(16.04 LTS).
In order to successfully build DeSpeect, The following step must be taken :
- Clone the Despeect repository
- Install Speect
- Configure the application
- Open a terminal
- Go to the <install-directory> (of your choice)
- Issue the command:
git clone https://github.com/TheBlackCat-SWE/Despeect.git
- Open a terminal
- Go to the <install-directory>/Despeect/SpeectLib/
- Issue the command:
install.sh
The compilation time may vary.
At the time of the writing, it is necessary to configure Despeect with QtCreator. The following step must be taken in order to correctly configure DeSpeect:
- Open QtCreator
- Select File -> Open File or Project
- Go to the directory <install-directory>/Despeect and open the file CMakeLists.txt
- Click on button 'Open'
- Click on button 'Configure Project'
In order to successfully generate a HRG the following step must be taken :
- Load a configuration file
- Insert text
- Select Utterance Type and Utterance Processor
- Generate HRG
- Select File -> Open File or Project (1)
- Choose the configuration file voice (you could find some inside the directory <install-directory>/Despeect/voices)
- Click on Open
Insert required text inside the text box (2)
Once loaded a configuration file, the 'Flow Control' widget(3) will contain a checkbox list of all Utterance Type and Processor available. Select the required utterance types and processor.
In order to generate HRG you can:
- Run All(run all Utterance processor inside the selected Utterance Type)
- Run Step (run Utterance processor individually)
clicking the button 'Run All' and 'Run Step' respectively.
The HRG will be shown in (5).