A list of Free Software network services and web applications which can be hosted on your own servers. With repository stars and repo forks.
I don't know what this has over the other awesome-selfhosted directory. Need to compare them.
Transcription of calls from trunk-recorder using OpenAI Whisper.
- transcribe.sh runs from trunk-recorder which makes a POST request to the API, passing along the call WAV and JSON
- API creates a new task to transcribe the call and adds it to a queue (in RabbitMQ)
- Worker(s) (running on a machine with a GPU) picks up the task from the queue and executes it, transcribing audio
- As part of the task, worker makes an API call to Meilisearch to add the transcribed call to the search index
- If notifications are configured, as part of the task the worker will send appropriate notifications
If you're using OpenAI Whisper, you can use a local GPU to accelerate computations.
Stract is an open source search engine where the user has the ability to see exactly what is going on and customize almost everything about their search results. It's a search engine made for hackers and tinkerers just like ourselves. No more searches where some of the terms in the query arent used, and the engine tries to guess what you really meant. You get what you search for.
Oh, and if we ever become evil (maybe by changing our motto) please take our code and start a competitor. The fact that you have this ability will make sure that our values will always be aligned with our users.
We will also have a paid API for developers.
The build process is outlined in CONTRIBUTING.md. It mentions an optional configuration option for Alice, an AI search assistant.
REST API docs: https://trystract.com/beta/api/docs/
SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled. It's a fork of the original Searx but has diverged in some important ways. Theming aside, it's more mobile friendly, the translations are up to date, preferences (such as "which engines are reliable on this instance") have been updated, and metrics for detecting broken or failing search engines supported by the software have been added (opt-out). No additional filters or proxies are needed.
I would think that it can be extended in the same ways as Searx but haven't explored it yet.
Minimalistic web-application for creating electronic music with virtual analog synthesizers. Initially designed as sketching tool but can be used as basic DAW. Simple and easy to use. Mobile-friendly. Client only, no data processed on servers. Subtractive synthesizers with various effects. Multi-layered step sequencer. Exportable synth presets. WAV audio export.
A collection of tools, calculators, converters, generators, encoders, decoders, and more of common use to IT. Can be built by hand but it's probably easier to download the latest release and unpack that.
MediaCMS is a modern, fully featured open source video and media CMS. It is developed to meet the needs of modern web platforms for viewing and sharing media. It can be used to build a small to medium video and media portal within minutes.
It is built mostly using the modern stack Django + React and includes a REST API.
Designed for personal use only. If you want people to be able to create accounts, you can do that. If you want it to be "everybody but me can only watch thigns," you can do that. If you want the entire install to be private and requiring logins from everybody, you can do that, too.
The open source password manager for teams.
This repository is a chat example with LLaMA models running on a typical home PC. You will just need a NVIDIA video card and some RAM to chat with model. By using HF version you may fine-tune the model to any desired task.
System requirements: A modern enough CPU, NVIDIA graphics card (2 Gb of VRAM is ok); HF version is able to run on CPU, or mixed CPU/GPU, or pure GPU, 64 or better 128 Gb of RAM (192 would be perfect for 65B model).
This repository is intended as a minimal, hackable and readable example to load LLaMA (arXiv) models and run inference by using only CPU. Thus requires no videocard, but 64 (better 128 Gb) of RAM and modern processor is required. Make sure you have enough swap space (128Gb should be ok :).
GPT4All is an ecosystem to train and deploy powerful and customized large language models that run locally on consumer grade CPUs. The goal is simple - be the best instruction tuned assistant-style language model that any person or enterprise can freely use, distribute and build on.
A GPT4All model is a 3GB - 8GB file that you can download and plug into the GPT4All open-source ecosystem software. Nomic AI supports and maintains this software ecosystem to enforce quality and security alongside spearheading the effort to allow any person or enterprise to easily train and deploy their own on-edge large language models.
Run any GPT4All model natively on your home desktop with the auto-updating desktop chat client.
A personal Mycroft backend. Alternative to mycroft.home, written in Flask. This repo is an alternative to the backend meant for personal usage, and allows you to run without using the Mycroft servers. There are no user accounts! This is NOT meant to provision third party devices, but rather to run on the Mycroft devices directly or on a private network.
llmapi-server is an abstract backend that encapsulates a variety of large language models (LLM, such as ChatGPT, GPT-3, GPT-4, etc.), and provides simple access services through OpenAPI.
Quay is a self-hosted Docker container registry. Supports Docker registry protocol v2, Docker manifest schema v2.1 and v2.2, image discovery and squashing, third-party authentication, and more.
Sequencer64 is a progressive web application that allows you to quickly sequence a 64-step pattern for a 9-sound sampler. It has 9 kits, each with 9 samples in which you can individually edit the pattern, 'slice' (sub-16th note subdivisions), pitch, length, and velocity. There are 9 pre-programmed sequences to get you started. Login with social media to save and load your sequences for later work. Share your sequences publicly with other users. Multiple color themes and a visual spectrum analyzer.
Live demo: https://mermaid.live/
A single file, self hosted, backend as a service.
Install. Create a desolid schema file. Run it and it exposes a GraphQL REST API. Uses just about any common database engine (including SQLite) as its back-end. You'll have to rebuild every time you add or update a schema.
Seems like it's in shark mode.
Create a ChatGPT like experience over your custom docs using LangChain. This repo can help you use models hosted on HuggingFace for embedding and for text generation.
Fraidycat is an app for Linux, Windows or Mac OS X which can be accessed from a local browser or a Tor onion site - and is a tool that can be used to follow folks on a variety of platforms. But rather than showing you a traditional 'inbox' or 'feed' view of all the incoming posts - Fraidycat braces itself against this unbridled firehose! - you are shown an overview of who is active and a brief summary of their activity.
Fraidycat attempts to dissolve the barriers between networks - each with their own seeming 'network effects' - and forms a personal network for you, a personal surveillance network, if you will, of the people you want to monitor. (It's as if the Web itself is now your network - imagine that.)
There are no fancy algorithms behind Fraidycat - everything is organized by recency. (Although, you can sort follows into tags and priority - "do I want to track this person in real-time? Is this a band that I am only interested in checking in on once a year?") For once, the point isn't for the tool to discern your intent from your behavior; the point is for you to wield the tool, as if you are a rather capable kind of human being.
Miniature, self-hosted chat server.
4996 links, including 376 private