Enabling the Translation of Local African Languages using Machine Learning

Local Chapter Accra, Ghana Chapter

Coordinated byGhana ,

Edem Tengey

Status: Completed

Project Duration: 23 Jan 2023 - 01 Apr 2023

Open Source resources available from this project

Github Repository

Project background.

Africa has more than 3000 Languages. There is no proper mechanism for the efficient and accurate translation of these languages into more common, vastly spoken global languages like English, French, and Spanish. I aim to use AI and ML techniques to provide translations for these languages.

The problem.

We want to solve the inadequate availability of translation technology for African languages. We believe this will greatly help with Africa’s interactivity with the rest of the Globe. This challenge in particular seeks to help with the translation of Ghanaian Languages.

Project goals.

1. Source for train datasets. 2. Carryout data preprocessing 3. Develop a Machine learning model  4. Carry out inference with the trained model using test data 5 Develop an API to be used by individuals.

Project plan.

  • Week 1

    Kick-off and Collection of Data

  • Week 2

    Data Cleaning and Exploratory data analysis

  • Week 3

    Feature engineering and selection

  • Week 4

    Build Machine Learning Model

  • Week 5

    Deploy through various means for use across different platforms

Learning outcomes.

1. Collection of Data.
2. Data Cleaning.
3. Data Analysis.
4. Build Machine Learning
5. Develop API

Share project on: