Resources / Top Repos

openai/whisper

Open-source speech recognition that turns audio into text and subtitles.

Whisper is an open-source speech recognition model from OpenAI. It transcribes audio in many languages, can translate speech into English, and runs on your own computer or server through Python, so recordings never have to leave your machine.

Good for

  • Transcribing interviews, meetings or training sessions into searchable text
  • Creating subtitle files for videos and social reels
  • Turning voice notes from customers or field staff into text an automation can process
  • Drafting minutes from recorded school or board meetings

Good to know

Whisper is free under the MIT licence. It runs much faster on a GPU, and the larger models are slow on an ordinary laptop. Accuracy varies by language and accent, so test it on your own recordings, including any Swahili or mixed-language speech, before relying on it.

Works well with: Python, n8n and remotion-dev/remotion.

More like this

Open source

laravel/laravel

The application skeleton that every new Laravel project starts from.

#laravel #php #starter

Open source

n8n-io/n8n

The source code for n8n, the workflow automation tool you can self-host.

#automation #typescript #self-hosted

Open source

filamentphp/filament

Build admin panels, forms and data tables for Laravel apps quickly, in PHP.

#laravel #admin #livewire

Trending Open source

ollama/ollama

Download and run open-weight AI language models on your own computer or server.

#ai #local-llm #models

You're offline. Some things won't load until you reconnect.
Chat on WhatsApp

SparkWorld

+254 110 757 810

Chat with us on WhatsApp

Before we open WhatsApp, leave your name and number.

Why we ask

  • On a computer, WhatsApp only opens if WhatsApp Web is linked to your phone, and you may not be logged in.
  • So we don't lose you: if the chat doesn't start, we'll message you first.
  • We only use your details to reply to you.
Skip and open WhatsApp now