Cards / needs / Image, audio, and speech AI

Image, audio, and speech AI

Self-hosted alternatives to Midjourney, Otter.ai, ElevenLabs and Descript transcription.

What people say

• generate images on my own GPU
• transcribe meetings privately
• text to speech without a subscription
• use my gaming PC's GPU from my laptop
• run Stable Diffusion or ComfyUI from anywhere

Why people self-host it

People leave Midjourney, Otter.ai, ElevenLabs, Descript transcription for it. Why, strongest first: cost, privacy, control.

Does it suit a machine of your own?

Fits, with conditions needs a GPU or Apple Silicon; it fits well when the person already has one at home

Questions that decide it

• what GPU or Mac do you have?
• images, transcription, or voice?

How to publish it

Usually

org: tools such as ComfyUI have no login of their own, so never none; an API for other apps on api-key

When the data is private or sensitive

org only, or not published

The machine

a GPU with plenty of memory, or Apple Silicon; lots of disk for models

Cards for it

No card yet. awesome-selfhosted's GenAI and Media Management categories, and GitHub repositories tagged comfyui, stable-diffusion, whisper, or text-to-speech An agent with the edgible-cards skill can shortlist apps and build one, or you can ask for one.