Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

KevKibe/Chat-based-Interactions-with-Swahili-Audio-and-Videos

Domain:

natural language processing

Record type:

software
Creator:
Kev
Host:
This is an application that enables a chat-based conversation in English with Swahili videos/audio files using Python and deployed as a REST API using Flask containerized using Docker. ## Description This is an application that enables a chat-based conversation in English with Swahili videos/audio files using Python. The application also works for other languages but I am farmiliar with Swahili hence why I picked the language. This is made possible by: - **Pytube** which is a library used to retrieve audio files from Youtube Videos. - **Whisper API** from OpenAI which is used to transcribe Swahili audio files to Swahili text. - **Google Translate API** to translate Swahili text to English. - **Langchain and OpenAI** enables the application to generate dynamic responses and engage in chat-based conversations in English. ## Limitations - Takes a while to transcribe Swahili audio files to text specifically 2mins for a 9min-long audiofile and 13mins for a 46min-long audiofile as shown in the notebook. - Google Translate API for translation is not that accurate. - OpenAI API has a token and request limit and for free trial users. It would be better to use a paid account. - The transcription part of the application requires alot of computation. I used a Tesla P100 GPU and it was running at maximum when transcribing the 46min-long audio ## Usage You can clone this repository and follow these steps or use a notebook as demonstrated here. - Ensure you have Python installed on your machine (version 3.6 or higher). - Install the required dependencies by running the following command: `pip install -r requirements.txt` - Run this command in your terminal `pip install git+github.com ` to get whisper by OpenAI - Install ffmpeg using instructions from this site - Set up environment variables: Create a `.env` file in the root directory of the project and add your OpenAI API key as follows: `OPENAI_API_KEY=your_api_key_here` - Run the app by running the command `python main.py` in your terminal which will prompt you with the youtube URL input and then proceed to ask for a prompt. - The application will transcribe the Swahili audio, …

Visit

github.com

Tasks

automatic speech recognitionspeech processing

Languages

Swahili

Similar

KevKibe/Finetuning-WhisperSmall-LoRA-SwahiliDiscovering Human Interactions in Videos with Limited Data LabelingKevKibe/Inflation_in_Kenya_TimeSeries_Predictionnileagi/swahili-chat-afriqueqwen35-4b-loraKevKibe/African-WhisperSwahili Audio Classification

KevKibe/Finetuning-WhisperSmall-LoRA-Swahili

This repository features code for Swahili Automatic Speech Recognition (ASR) fine-tuning using the W

Discovering Human Interactions in Videos with Limited Data Labeling

We present a novel approach for discovering human interactions in videos. Activity understanding tec

KevKibe/Inflation_in_Kenya_TimeSeries_Prediction

This is a time-series forecasting application used for predicting future inflation rate in Kenya usi

nileagi/swahili-chat-afriqueqwen35-4b-lora

KevKibe/African-Whisper

🚀 Framework for seamless fine-tuning of Whisper model on a multi-lingual dataset and deployment to p

Swahili Audio Classification

Can you classify Swahili audio into words?
The data was collected by 300 taskers from Kenya. There are twelve classes, each a different audio utterance in Swahili.
The objective of this competition is to classify the 12 different Swahili words, using mac