Emotive Speech Python Package

A Standard Python Package For Emotive Speech Generation Based on DAVID: An open-source platform for real-time emotional speech

transformation using pysox

Pre-Requisites

pysptk

A python wrapper for Speech Signal Processing Toolkit (SPTK) https://github.com/r9y9/pysptk | sudo pip install pysptk

Basic Signal Processing Tools

Scipy,Numpy Make Sure scipy.io- dedicated wavefile read/write library- is also installed

pysox

a Python library that provides a simple interface between Python and SoX. we will be using the Transformers (sox.transform.Transformers) for synthesis. http://pysox.readthedocs.io/en/latest/api.html | sudo apt-get install pysox

Installation and Setup

After fulfilling all the requirements,The Python Package of the Emotive Speech Project can be cloned as:

$ git clone https://github.com/dergkat/emotivespeech.git

Usage

$ cd ESPP/src
$ python -B EmotiveSpeech 
usage: emotivespeech.py [-h] [-c chunk_size] [-s semitones] [-r cutfreq]
                        [-g gain] [-q qfactor] [-v speed] [-d depth]
                        [-o tempo] [-i intensity] [-p parameter_control]
                        filename typeOfEmotion

Example

$ python -B emotivespeech.py /home/user/Desktop/TestFolder/Test.wav sad

Arguments

$ python -B EmotiveSpeech.py arg1 arg2

arg1(Positional Argument): Absolute Path For Wavefile

arg2(Positional Argument): TypeofEmotion: (happy,sad,afraid,happy_tensed)