Free local tools: chat, voice, transcription and images

What you'll see in this lesson

1. Your free AI setup at home

   With free programs you can set up chat, transcription, voice and images that run on your machine, with no subscriptions.

2. A server that speaks like OpenAI

   Ollama, LM Studio, Jan and LocalAI offer the same API as OpenAI on your machine: you only change the base address.

3. Open WebUI: a private ChatGPT in your browser

   Open WebUI is a web interface that connects to Ollama or any compatible API and adds documents, voice, images and users.

4. Jan and LM Studio: desktop apps, no terminal

   If you prefer a regular app, Jan and LM Studio download models, chat and also start a local server.

5. LocalAI: one API for text, voice and image

   LocalAI is a free, MIT-licensed server that mimics the OpenAI, Anthropic and ElevenLabs APIs for all kinds of models.

6. Transcribe audio locally with Whisper

   Whisper, OpenAI's open model, transcribes meetings and audio in Spanish on your machine; whisper.cpp and faster-whisper speed it up.

7. Natural voice locally: Kokoro and Piper

   Kokoro and Piper turn text into speech on your machine, for free and with Spanish voices, without sending the text to any service.

8. Images on your machine with ComfyUI

   ComfyUI generates and edits images and video locally with models like FLUX.2, Qwen Image or Wan, by linking blocks called nodes.

9. Summary: free local tools

   With free pieces that speak OpenAI's language you get chat, transcription, voice and images with no subscriptions and without your data leaving.