8000 GitHub - sig5/mood-synth: Speech Synthesis based on Mood analysis of RAVDESS dataset
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

sig5/mood-synth

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
.idea
.idea
 
 
venv
venv
 
 
README.md
README.md
 
 

Repository files navigation

mood-synth


mood-synth is a speech synthesis GUI based python program which predicts the mood of user with a Machine learning Model based on RAVDESS dataset.

How to use?


  • Execute "/venv/bin/ui.py" and appropriate GUI will be launched.

    Features

  • Uses RAVDESS dataset licensed under Creative Commons License.
  • Classifies Data on the basis of MLP Classification.
  • Uses reLU function as the corresponding activation function.
  • An accuracy of around 70%-78% has been recorded.
  • A Tkinter based GUI interface that provides user ease of accesiblity.
  • Modern Material Design

    • Input


    Input is currently provided in .wav file format that is to be chosen using GUI interface provided in th ui.py file included in the /bin folder

    Dataset


    The The Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) is being used as the Dataset for model training. The sample rate of the Dataset file has been reduced to lower the file-size.

    Output


    The data is classified into four categories as of now
  • Calm
  • Happy
  • Fearful
  • Disgust

    • References


    https://zenodo.org/record/1188976#.Xo_ww-nhXaI

    About

    Speech Synthesis based on Mood analysis of RAVDESS dataset

    Resources

    Readme
    Activity

    Stars

    2 stars

    Watchers

    2 watching

    Forks

    0 forks
    Report repository

    Releases

    No releases published

    Packages

    No packages published

    Languages
    0