cross-posted from: https://lemmy.world/post/45721955

cross-posted from: https://lemmy.world/post/45721951

cross-posted from: https://lemmy.world/post/45721900

cross-posted from: https://lemmy.world/post/45721589

Hi All, It has been while,

Dograh is an open-source, self-hostable voice AI agent platform. Think n8n but for phone calls. Visual workflow builder, inbound and outbound calling, bring your own LLM, STT, and TTS.

GitHub: https://github.com/dograh-hq/dograh

Setup

one command with Docker, about 2 minutes. No signup or API keys needed to get started:

What is new

Pre-call data fetch. Hit your CRM, ERP, or any HTTP endpoint during call setup and inject the response into your prompts. The agent greets the caller by name, references their account status, skips the "can I get your customer ID" step. Configure a POST endpoint in the Start Call node - API key, bearer, basic, or custom header auth supported. 10-second timeout; if the endpoint fails, the call continues without the extra context. Reference fetched values anywhere in prompts with {{customer_name}} syntax.

Pre-recorded voice mixing. Drop in actual human recordings for the predictable parts - greetings, confirmations, hold messages - and let TTS handle only what needs to be dynamic. The greeting sounds human because it is. Latency goes down, TTS costs go down.

Speech-to-speech via Gemini 3.1 Flash Live. One single streaming connection replaces the separate STT, LLM, and TTS hops. Turn response latency drops noticeably and the conversations feel more natural.

Post-call QA with sentiment analysis and miscommunication detection. Full per-turn call traces via Langfuse.

Tool calls, knowledge base, variable extraction are all there too.

What is coming

Real-time noise separation for live call streams - still the thing I most want to solve after last week's thread. BSD-2 licensed.

GitHub: https://github.com/dograh-hq/dograh

Special thanks to this community that supported me with my last post ❤️

Happy to get feedback and contributors. A star would mean a lot

---

I'm putting together a bunch of my old PC parts to get a home server going, and i was wondering what the best OS would be. I have Mint installed on it currently, but I'm not sure thats the best option so I'm looking for alternatives.

I'm definitely a Linux novice, so I dont want anything super complicated. I was thinking of using Debian since it seems decent, but I really know nothing about the different distros and would appreciate a good starting point if anything

I have a Syn DS 220+ 2Gb RAM and a GmkTec miniPC with Intel N100, 16 GB DDR4 running Proxmox V8.4...

NAS services:

• Synology Office + calendar
• Restic
• Rsync

Proxmox services:

• piHole
• Jellyfin + related services
• Docker with less than 5 stacks.

The miniPC is at its limits with NVME and sometimes even piHole hits a snag.

Recently I finished helping a friend setup TrueNAS 25.04 and I love it.

I could get my hands on a SFF PC at 180$ and some 64 GB DDR4 ECC with 90$.

I would like to also host Forgejo and maybe the frontend of some small website. Ideally, I'd like to setup Kerberos for NFS sharing too and with Synology that failed spectacularly.

So, after all this context, would you advise me to migrate from Syn+Nucbox to TrueNAS or not? What is your reasoning?

Trying to figure it out and work on extensions but it seems their own plugins can’t work on your server as extensions…?

Also, would be curious to hear of anyone else’s TRMNL experiences (beside the cringe lord CEO).

Self-hosted hosting control panel using Cloudflare Zero Trust Tunnels to securely route multiple domains from a single machine, even on a residential ISP without opening firewall ports. Includes SSL, Multi-PHP, PHPMA, DB, DNS, Backups, WireGuard management and more.

https://inetpanel.info/

Hi there, i have a few questions about GL-Mt3000 and Openwrt.

Context Wall fell free to skip:

Questions:

Is there any benefit to host AdguardHome outside the router? I did it to learn, but i don't know if it has any advantages.

I plan to learn openwrt and flash the router to vanilla openwrt. My reasons are that:

• I feel restricted by the gl-inet interface.
• Gl-inet doesn't seem to update too frequently their firmware.
• I'm worried their custom software will cause problems if i tinker Luci too much.
• I think it will easier to learn the vanilla version than a custom version.

Does all of this make sense? Do you think is worth to spend time on this?

New release: Newspipe 11.3.0 🚀

Security & privacy take the spotlight in this version:

• Fixed multiple XSS and SSRF vulnerabilities (thanks fyrepaw13 🙌)
• Safer API with stricter field validation and sanitization
• State-changing routes now protected with POST + CSRF tokens
• More privacy-friendly bookmarks page

Plus UX improvements across bookmarks, forms, and charts.

https://github.com/cedricbonhomme/newspipe

I built ActuFeed because I wanted a simple, distraction-free way to follow exactly the news sources I care about — without algorithms, ads, or tracking.

Key features:

• Fully customizable tabs and feeds (add any RSS or website) Smooth bilingual interface
• Clean reading experience optimized for desktop (works on mobile too)
• Very lightweight and easy to self-host with Docker
• No account required

You can try it instantly here: → https://actufeed.com/

GitHub repo (open source + Docker): https://github.com/drenlia/actufeed

Would love your feedback or suggestions!

A very jovial greeting to all,

About 20 minutes ago, I started the build for Hubzilla 11.2; as usual, it will be available for all to enjoy, and update their own instances after about an hour or so, so please don't update until then.

If you're curious about the code, you are most welcome to check out the Hubzilla code at: https://framagit.org/hubzilla/core/-/releases

and, of course, the docker image code at: https://github.com/dhitchenor/hubzilla

Questions, issues and PRs are all welcome; I'm looking forward to speaking with you.

نُشر تبادليًا من: https://hexbear.net/post/8061686

We have been without support for over five days. My children and I are struggling, and we desperately need help to survive. Every little donation can make a huge difference in our lives.

💙 Please, do not forget us. Your kindness can bring hope, food, and safety to my family.

Donate here: https://gofund.me/00439328

I upgraded from a thinkpad x220, i5 2520M and 4gb of ram to my old desktop computer, with a FX6300, 16gb of ram and a GT 1030.

I already run my own mbin instance in there (the reason for the upgrade), but as it's a lot faster than the instance needs it to be, I would like to host other stuff too, mainly to Brazil.

I want to host things like searxng, invidious, maybe redlib too, but idk about other services. As there's the GPU in there, there could be some transcoding too.

Maybe a mixture of private and public services.

Hi guys, I wanted to share a project I’ve been working on called xSpa. It's an implementation of Single Packet Authorization that works at the XDP level.

I built this because I wanted something faster and more DDoS-resilient than traditional port-knocking or SPA tools that rely on userspace processing or iptables. Here, the "drop-all" logic happens right at the driver level.

Key bits:

L1 verification (SipHash) in kernel space.

L2 (ChaCha20-Poly1305) in Go userspace.

It uses the eBPF ring buffer for communication.

This is my first Go project and my first shot at Open Source. I’m still a bit of a noob when it comes to kernel-level programming, so I’d love to get some feedback on the architecture and security. If anyone has time to check the code, I’d love to hear your thoughts on how to make it better.

نُشر تبادليًا من: https://hexbear.net/post/7952295

نُشر تبادليًا من: https://hexbear.net/post/7952294

نُشر تبادليًا من: https://hexbear.net/post/7952199

I am Abeer, a mother of four children from Gaza. We went through very hard times and lost everything because of the war. These days we are simply trying to survive and meet the most basic needs.

Today we only need $100 so I can buy some essential things for my children like food and milk.

Any help, even a small donation, can make a big difference for us. And if you cannot donate, please share our story so it can reach someone who might be able to help Link in bio or message me privately. 🙏💔🇵🇸

cross-posted from: https://lemmy.ml/post/44496833

4 years ago, I posted about the weather service I did build. Back then, Serenum used OpenWeatherMap, but after a while they changed their API and I didn't have any strength to make the changes in Serenum API.

Earlier this year, I started with the new version of Serenum without Serenum API (too much work, too little time), now using Open-Meteo. I released the first version last week as a beta and now I find Serenum enough finished to share it here on Lemmy.

Still in beta, though, since stuff needs to be improved. But everything(?) works as it should.

Voiden is an offline-first, git-native API tool built on Markdown Voiden is an API client we have been building that takes a different approach from most existing tools.

It didn’t start with the idea of “building a better Postman”.

A bit of background. Over time, API tooling has become heavyweight: cloud dependencies for local work, forced accounts, proprietary formats, and workflows that break the moment you are offline. On top of that, time wasted on fixing API specs that don’t match the code, docs in separate random tools, tests also separate and an overall governance mess. Not to mention collaboration.

So we asked a simple question: What if an API tool respected how developers already work?

That led to a few core ideas:

• Offline-first , no accounts, no telemetry
• Git as the source of truth.
• Plain text files: specs, tests, and documentation live together in Markdown
• A programmable interface instead of static forms: requests are composed from reusable blocks (endpoints, headers, auth, params, bodies, etc.) that you can structure the way you want
• Plugin system for extending functionality rather than bloating the core with new features Some of our core plugins include gRPC,GraphQL,WebSockets,etc…

We have just also updated our docs to welcome community plugins, so teams can extend the tool for their own workflows or integrations. https://docs.voiden.md/docs/plugins/build-a-plugin

We opensourced Voiden because extensibility without openness just shifts the bottleneck. If (API) workflows should be transparent, the tools should be too.

Welcome to try out and share feedback- happy to chat with everyone.

Strong opinions are encouraged. :)

Github : https://github.com/VoidenHQ/voiden

Download here : https://voiden.md/download

What tools and/or resources can I use to start such a project?

I've been putting some research into it but besides getting conflicting views, I always get the feeling the advices being offered feel exaggerated for what I have in mind.

My intention is to solely publish text.

I was thinking on getting a RPi, perhaps an older model, to use it as host. My idea is to have the public facing machine completely isolated from my daily use computer and network, so that in the event the site gets attacked I can simply re-upload everything and be done.

For the website creation itself I've been thinking about using Hugo to start but in the background learn some HTML and build a simple site by myself to replace the original.

What am I not considering or misjudging here?

Are there any other tools I should be looking into?

Any thoughts on this are welcome and apreciated.

How I could get a #LibreELEC Image into a #LXC or #VM container and AutoStart it on boot without interaction?

In best case without the need that this container use the graphic core exclusively.

I have to use one device for several usecases, but I do wish to split my #Mediacenter from other #SelfHosting projects.

#helpneeded #LinuxServer @selfhost @docker@lemmy.ml

Hi, not sure where else to post this on the Fediverse, if there is a better community let me know and I will move this post there :)

I recently was trying to figure out an easy way to negate spam calls for my family. I am aware of tools like Spam Blocker however they are not fool proof. I don't really get spam calls myself, but the rest of my family does. The reason I suspect this is, is because I do not give out my phone number unless it is strictly necessary. I do not sign up for store memberships, when I'm on online shopping websites I usually just provide a dummy non-working number, and the only time I really do give my phone number out is in cases of emergency or for close family and friends.

This made me think, what if the family had a shared number to give out to take advantage of store memberships, online shopping, and whatever else. Then I could have everyone rotate their current numbers and do what I do where they only give out their main phone number in case of emergency. Being a self-hoster, this then evolved from a "family phone" that is kept at home, to "what if I could utilize SIP technology to let anyone in the family access the number from their own phones". I've tried looking into this on my own, but the most I can find is people with similar ideas abandoning the idea in favor of true telephony/SIP services.

The reason a telephony service won't work is a lot of the businesses have gotten smart about filtering phone numbers and will not let you sign up with one of these services (Especially if they have an old-fashioned 2FA method like SMS.) Not to mention, I prefer paying for my phone service in cash at a physical store for the added privacy of not having to give a real name.

What I was thinking was SMS USB Modem -> Home Lab -> SIP Program(?) -> SIP Clients (Phones).
Is this possible? Would SMS/MMS/Voice calls work? Is there a better solution that would be easier (While still working for my use case of using a real SIM card?)

Thanks in advance, hope you all are having a lovely day :)

I saw some projects, but they all seem not really polished. It would be nice to be able to share sound too.

cross-posted from: https://lemmy.world/post/42574918

I am getting started with self hosting and one of the things I would love to host is a Signal TLS proxy using Docker.

Problem is that I have ports 80 and 443 taken by Nginx Proxy Manager (also in a Docker container), through which I forward to different services depending on the subdomain.

I tried modifying the docker-compose.yml file to use ports 9443 and 980 and have it working using a certificate created on NPM, but to no avail.

Being a beginner, it can well be that I don't understand reverse proxies well enough, but that's why, with your help I would love to take this opportunity to learn more.

Thanks in advance.

cross-posted from: https://lemmy.ml/post/42502523

Charmarr aims to make self-hosting a media stack in Kubernetes easy. It provides charmed versions of *arr applications (radarr, sonarr, etc) and some friends. Charms are operational wrappers, that is they configure the underlying applications themselves. So deploying Radarr, Prowlarr, SABnzbd, Gluetun. Setting a required TRaSH profile Radarr and routing the traffic of SABnzbd and Prowlarr via Gluetun VPN tunnel can be done with a few intuitive commands. An example command snippet would be sth like

# deploy the apps into my cluster
juju deploy radarr-k8s
juju deploy prowlarr-k8s
juju deploy sabnzbd-k8s
juju deploy gluetun-k8s

# tune 4k TRaSH guide profiles in my Radarr
juju config radarr-k8s variant=4k

# Route Prowlarr and SABnzbd via Gluetun
juju integrate sabnzbd-k8s gluetun-k8s
juju integrate prowlarr-k8s gluetun-k8s

This can be extended to any cross-communicating tools like Overseerr (already part of charmarr), Plex (already part of charmarr), Huntarr (planned) etc.

This also enables the using OpenTofu to deploy the entire stack using a single command:

tofu init && tofu apply

okay, 2 cmds. This way your entire media stack can have a declarative deployment using a single 20-30 line .tf file instead of multiple manifests. This sets up all applications, handles storage, handles VPN routing (you just provide your VPN credentials and media paths), connects everything together, and it's ready in about 10 minutes. You just need to log into Plex, connect Overseerr, and add your indexers.

"But K8s is overkill for a homelab and no one needs it"

Totally agree. This is intended to make managing the media stack in a Kubernetes substrate easier for the veterans and loosening up the entry point into Kubernetes for a self-hosted media server for this who are interested in K8s. That said, it will still have some learning curve if you're completely new to K8s.

All the tools that are part of charmarr including charmarr itself (except for Plex, which I plan to switch with or add on top of Jellyfin eventually) are open source and free to use.

Is it stable?

It's been running in my lab for more than a month without any issues. But, I wouldn't call it stable yet especially if you enable all the fancy bells and whistles, but I've been running nightly tests deploying the stack using tofu and tearing it down and it's been consistently successful. If you're interested in experimenting or using it, its enough that you have a Ubuntu system (I also have an oneliner to setup the required infra to deploy charmarr).

Here's the repo - https://github.com/charmarr/charmarr

PS: sorry about the ads on the docs site. It's hosted by readthedocs and they include ads on the free version.

cross-posted from: https://discuss.online/post/34584845

Curious on what tools people would recommend, either from clients, locally or self-hosted.

• privatebin works nicely as a basic pastebin.
• stuffedanimalwar is just silly fun, with group drawing collaboration and chat that only exists in the active client session.

cross-posted from: https://discuss.online/post/34494723

Detailed episode for pairing with the very light "A Great Day for Linux". Hope you enjoy it. Since Lemmy struggles with markdown from Castopod, here is a link to the notes.