2026-05-16, 03:55 AM
(This post was last modified: 2026-05-16, 04:08 AM by sonicaaaa. Edited 1 time in total.)
Hey all,
I've been working on a Jellyfin plugin that creates a personal Alexa skill for your server. Each Jellyfin user gets their own skill with a customizable invocation name, so "Alexa, ask Jellyfin Player to play songs by Daft Punk" just works.
It actually builds on a the original work found here, https://github.com/infinityofspace/jelly...exa-plugin, that taught me that a Jellyfin plugin can create Alexa skills on my behalf.
What it does:
- Playback: play specific songs, albums, artists, videos, TV episodes, podcasts, and playlists by name
- Discovery: "play something new this week", "play happy music", "play 80s rock", "recommend something"
- Queue control: next, previous, shuffle, repeat, start over, add to queue, play next
- Favorites: "play my favorites", "I like this song", "remove from favorites"
- Media info: "what's playing?", "what's up next?", "continue watching"
- Radio mode: continuous playback based on your library
- Visual cards: album art and now-playing info on Echo Show/Spot devices
- Indexes your media db and sends it to the Alexa skill, so to increase the chances Alexa recognize the titles and names of your media and also add phonetics aliases to accomodate for language specific bad English pronunciation
Multi-language support is where I need your help:
The skill supports 17 locales across 11 languages. I've written custom utterances for all of them, but I'm a native Italian speaker ; my English, German, French, Spanish, Dutch, Arabic, Hindi, Japanese, and Portuguese voice commands were written by an AI. They work, but they might not feel natural to native speakers.
If you speak any of these languages, I'd love your feedback on whether the voice commands sound right or could be improved. The utterances live in simple JSON files in the repo ; PRs welcome, or just open an issue with suggestions.
Supported languages: English (US/UK/AU/CA/IN), Italian, German, French (France/Canada), Spanish (Spain/Mexico/US), Portuguese (Brazil), Dutch, Arabic, Hindi, Japanese
Getting started:
Install from the plugin repository in Jellyfin 10.11+
Create a free Amazon Developer account and set up a Security Profile
Configure the plugin, authorize with Amazon, link your account in the Alexa app
Start talking to your media server
IMPORTANT: It requires your server is exposed to the internet, accessible via https via DNS; a requirement imposed by Alexa servers to connect to it.
Repo: https://github.com/paoloantinori/jellyfin-alexa-plugin
It's alpha software ; things may break between releases. Back up your config before updating. Feedback, bug reports, and translation improvements are all very welcome.
(Cross posted on Reddit too)
I've been working on a Jellyfin plugin that creates a personal Alexa skill for your server. Each Jellyfin user gets their own skill with a customizable invocation name, so "Alexa, ask Jellyfin Player to play songs by Daft Punk" just works.
It actually builds on a the original work found here, https://github.com/infinityofspace/jelly...exa-plugin, that taught me that a Jellyfin plugin can create Alexa skills on my behalf.
What it does:
- Playback: play specific songs, albums, artists, videos, TV episodes, podcasts, and playlists by name
- Discovery: "play something new this week", "play happy music", "play 80s rock", "recommend something"
- Queue control: next, previous, shuffle, repeat, start over, add to queue, play next
- Favorites: "play my favorites", "I like this song", "remove from favorites"
- Media info: "what's playing?", "what's up next?", "continue watching"
- Radio mode: continuous playback based on your library
- Visual cards: album art and now-playing info on Echo Show/Spot devices
- Indexes your media db and sends it to the Alexa skill, so to increase the chances Alexa recognize the titles and names of your media and also add phonetics aliases to accomodate for language specific bad English pronunciation
Multi-language support is where I need your help:
The skill supports 17 locales across 11 languages. I've written custom utterances for all of them, but I'm a native Italian speaker ; my English, German, French, Spanish, Dutch, Arabic, Hindi, Japanese, and Portuguese voice commands were written by an AI. They work, but they might not feel natural to native speakers.
If you speak any of these languages, I'd love your feedback on whether the voice commands sound right or could be improved. The utterances live in simple JSON files in the repo ; PRs welcome, or just open an issue with suggestions.
Supported languages: English (US/UK/AU/CA/IN), Italian, German, French (France/Canada), Spanish (Spain/Mexico/US), Portuguese (Brazil), Dutch, Arabic, Hindi, Japanese
Getting started:
Install from the plugin repository in Jellyfin 10.11+
Create a free Amazon Developer account and set up a Security Profile
Configure the plugin, authorize with Amazon, link your account in the Alexa app
Start talking to your media server
IMPORTANT: It requires your server is exposed to the internet, accessible via https via DNS; a requirement imposed by Alexa servers to connect to it.
Repo: https://github.com/paoloantinori/jellyfin-alexa-plugin
It's alpha software ; things may break between releases. Back up your config before updating. Feedback, bug reports, and translation improvements are all very welcome.
(Cross posted on Reddit too)