# Feedbard

Feedbard is a self-hosted tool that turns your RSS and Atom feed articles into a narrated audiobook library for Audiobookshelf. It is open source and self-hosted, installable on Unraid.

Canonical page: https://ourlocalhost.com/app/6792/feedbard

## Description

Feedbard is a self-hosted tool that turns your RSS and Atom feed articles into a narrated audiobook library for Audiobookshelf. It polls your feeds, fetches full articles, translates them into your chosen language, describes images and charts for speech synthesis, and publishes each article as a tagged MP3 episode with per-article cover art. Designed for personal listening, it runs on a configurable schedule, tracks processed posts in a local database, and never re-narrates the same article after a restart.

## Platform availability

Available on 1 of 10 platforms: Unraid.

## Health score

47/100 (Fair), based on 3 of 4 factors.
- Maintained: 83 (last commit 64 days ago)
- Popular: 0 (0 GitHub stars)
- Easy to install: 35 (1 install platform, install notes, Docker image)

## Links

- Unraid: https://ca.unraid.net/apps/feedbard-1qmolmd1jymwxt?q=feedbard
- GitHub: https://github.com/Theviki20110/feedbard

## Common questions

### Where do I add the feeds Feedbard processes?

The feed list is at FEEDS_LIST_PATH, default data/feeds_list.txt. It ships empty, so add RSS or Atom feed URLs before the container processes anything. Only include feeds whose terms permit derivative use.

### Will Feedbard repeat an article after a container restart?

It records processed URLs in a local sqlite database and drops entries already seen. That prevents re-narrating the same article after a restart.

### How do I stop a large feed backlog from making one run very long?

Feedbard caps new entries to MAX_POSTS_PER_RUN on each cron tick. That keeps one run from converting a whole feed backlog at once.

### Does Feedbard require a cloud LLM or ASR service?

It supports Anthropic, Bedrock, or Ollama for LLM, and its ASR step is optional. Leaving ASR_BASE_URL empty skips transcription-based QA and halves the per-block cost.

### What restrictions apply to the generated audio?

The output is a derivative work, so do not redistribute it. That includes exposing the Audiobookshelf library holding it to anyone but yourself. Paid or subscriber-only posts should stay off the feed list unless the publisher's terms say otherwise.

---
Source: Our Local Host (https://ourlocalhost.com/app/6792/feedbard), the self-hosted and open-source app directory.
