# EDDI: the Elite Dangerous companion that speaks, logs and relays your journal

> EDDI (the Elite Dangerous Data Interface) turns the game's journal events into speech, VoiceAttack variables and uploads to EDSM, EDDN and Inara. It is a Windows application, and the interesting part is not the speech but the monitor and responder pipeline underneath it.

**EDCD/EDDI** — Companion application for Elite Dangerous

- Repository: https://github.com/EDCD/EDDI
- Stars: 525 · Forks: 92
- Language: C#
- License: NOASSERTION
- Published: 2026-09-17 · Updated: 2026-09-17 · Language: en
- Canonical page: https://hysenlabs.com/projects/edcd-eddi

## What EDDI does that the game does not

Elite Dangerous writes a journal of what happens in your session, but the game never tells you about it in a useful voice. EDDI is the layer that reads those events and reacts. The README describes it as a companion application providing responses to events that occur in-game, using data from the game and from third-party tools. The scope is wider than speech: the same event that produces a spoken line can also push a jump to EDSM, forward market or exploration data to EDDN, or set a variable that a VoiceAttack script reads. That combination is the reason people install it. A player who only wants text-to-speech has simpler options; a player who wants one process to own the journal, the speech and the third-party uploads is the target user. The repository topics list the integrations explicitly: eddn, edsm, frontier-api, text-to-speech and voiceattack-plugin.

## Monitors, responders and the event flow

The architecture is a three-stage pipeline, and the README spells it out. Monitors are pieces of code that check for information, for example a new Galnet article being published or an entry in Elite's journal. They hand the event to EDDI. EDDI then carries out operations based on the event: if the event says the player has changed system, EDDI fetches updated system information. Once the required information has been gathered, EDDI notifies each responder. Responders act on the event plus the gathered data. The EDSM responder sends details of jumps to EDSM so the player keeps a permanent flight log. The speech responder produces a spoken response. The VoiceAttack responder exposes variables to VoiceAttack scripts and runs specific scripts when events occur.

The repository layout matches that description: JournalMonitor, GalnetMonitor, StatusMonitor, ShipMonitor, MissionMonitor, MaterialMonitor, CrimeMonitor, CargoMonitor, CommanderMonitor, FleetCarrierMonitor, NavigationMonitor and EDDPMonitor on the input side, with SpeechResponder, EDSMResponder, EDDNResponder, InaraResponder and VoiceAttackResponder on the output side. Monitors and responders can be enabled or disabled individually from the UI, which is the practical control surface: if one integration misbehaves you can switch that responder off without losing speech or logging. Note the asymmetry in the README's own example: it says system information is fetched from EDDB, while the repository contains an EDDPMonitor and an EddiSpanshService, so the documentation is behind the code on which data source is used.

## Installing EDDI and getting a first spoken response

EDDI ships as a Windows installer from the GitHub releases page, and the README gives the default target directory. Download the current release, run the installer, and accept the default path unless you have a reason to change it: the README warns that if the installation is not a subdirectory of your VoiceAttack Apps directory, EDDI cannot be used as a VoiceAttack plugin. The path below is the documented default.

```text
C:\Program Files (x86)\VoiceAttack\Apps\EDDI
```

If you prefer to build it yourself, the README points at the sources and a separate dependency document rather than listing build steps inline.

```bash
git clone https://github.com/EDCD/EDDI
```

On first launch EDDI opens a window with a number of tabs. The README's instruction is to read each tab and configure it to taste, because each tab explains its own function. The speech responder is the one to check first: EDDI uses the standard Windows TTS voices, and a voice only appears in EDDI if Windows' TTS system can see it, which normally happens at install time. If a voice is missing from the list, the README says to check the Windows TTS settings rather than EDDI's own configuration. Once a voice is selected and the speech responder is enabled, the spoken responses come from the same event stream that feeds the other responders, so a single jump can produce a line of speech and an EDSM upload at once.

## VoiceAttack is optional, but the install path is not

The README is clear that the default install directory is fine whether or not you own VoiceAttack, which tells you the two products are decoupled at runtime but coupled at the filesystem level. The constraint is one-directional: install EDDI outside a VoiceAttack Apps subdirectory and you lose the plugin mode permanently until you reinstall. That is a strange design to inherit from a Windows installer, and it is the first thing to get right.

Version compatibility is the second. The README states that EDDI version 4.1.9 and below support VoiceAttack 1 rather than VoiceAttack 2. If you are on VoiceAttack 2 and running an old EDDI, the integration is not expected to work, and the fix is to move EDDI forward, not to reconfigure scripts. Initial integration is automatic; the deeper work, exposing your own VoiceAttack scripts to EDDI variables and triggering scripts on events, is documented on the VoiceAttack EDDI wiki page rather than in the README.

## Voices, SSML and where the speech quality actually comes from

EDDI does not ship a speech engine. It uses the Windows TTS voices, and the README recommends buying from Ivona or Cereproc if you want quality beyond the built-in set. The interesting detail is the reason given for preferring Ivona: those voices typically adhere more closely to published SSML standards and provide fuller functionality for phonetic speech, pauses mid-sentence and correct pronunciation of names such as ships, powers and systems. That matters for this application specifically, because Elite Dangerous is full of proper nouns that a generic voice will mangle. If pronunciation of system and ship names is the thing you care about, the voice purchase is the decision point, not EDDI's settings. There is also a Lexicons.md file in the repository, which suggests EDDI carries its own pronunciation lexicon, though the README does not describe how it interacts with a third-party voice's SSML support.

## Upgrades, uninstalls and the 1.x trap

The upgrade story splits by version, and the README does not soften it. Upgrading from EDDI 1.X is not an in-place operation: the recommendation is to uninstall the existing version and remove the %APPDATA%\EDDI directory before installing the new one, to get a clean installation and reduce the chance of problems. Upgrading from 2+ is handled by the installer, so running it is enough. That means the cost of a major-version jump is losing your local configuration and any custom speech or script work stored under that directory, unless you back it up first. The README does not document rollback, so there is no stated path back to a previous version if a new release breaks a responder you depend on.

Uninstalling goes through the Windows control panel, and the same %APPDATA%\EDDI directory holds everything EDDI creates. It can be removed on uninstall. If you are troubleshooting rather than leaving, the README routes you to TROUBLESHOOTING.md first and then to the issue tracker.

## Licence and maintenance signals

The repository's licence field is NOASSERTION, which means GitHub's detector could not classify the terms automatically; the repository carries both LICENSE.md and LicenseFile.txt at the top level. Treat that as unresolved rather than permissive. If you plan to redistribute EDDI, bundle it with your own installer, or reuse its code, read those two files and the ChangeLog rather than assuming an MIT-style grant. Nothing in the README states redistribution terms.

On activity, the last push to the default branch was on 2026-09-14, three days before this was written, and the most recent release is 5.0.5 from 2026-08-03. The repository is not archived. Development happens on the develop branch, with stable and beta branches carrying their own CI status badges, so the release cadence you experience depends on which branch you track. The practical upgrade cost is the installer run plus, in the worst case, the %APPDATA%\EDDI reset described above.

## Conclusion

Adopt EDDI if you play Elite Dangerous on Windows and want the journal to do something: spoken callouts, VoiceAttack variables, flight logs pushed to EDSM, EDDN and Inara. Skip it if you are on Linux or macOS, or if you want a tool that still works when Frontier changes the journal, because EDDI's monitors are tied to the game's event formats and the project ships fixes on its own schedule. Before committing, read the tab-by-tab configuration window, confirm your VoiceAttack major version, and check the troubleshooting page at TROUBLESHOOTING.md when a monitor goes quiet.

## FAQ

### What is EDDI?

EDDI, the Elite Dangerous Data Interface, is a companion application for Elite Dangerous that responds to in-game events using game data and third-party tools. Monitors pick up events such as journal entries or new Galnet articles, EDDI gathers related information, and responders produce spoken responses, VoiceAttack variables and uploads to services like EDSM and EDDN.

### How do I download and install EDDI?

Download the installer from the GitHub releases page at https://github.com/EDCD/EDDI/releases and run it. The default install path is C:\Program Files (x86)\VoiceAttack\Apps\EDDI, and the README notes that if the installation is not a subdirectory of your VoiceAttack Apps directory it cannot be used as a VoiceAttack plugin.

### Does EDDI work with VoiceAttack 2?

Yes for current versions, but the README states that EDDI version 4.1.9 and below support VoiceAttack 1 rather than VoiceAttack 2. Initial integration is automatic once EDDI is installed inside the VoiceAttack Apps directory.

### How do I upgrade EDDI from an older version?

From version 2 onward the installer handles the changes, so running it is enough. From version 1.X the README recommends uninstalling the existing version and removing your %APPDATA%\EDDI directory first to get a clean installation.

### Why can't I see my text-to-speech voice in EDDI?

EDDI uses the standard Windows TTS voices, and a voice must be visible to Windows' TTS system to appear in EDDI, which usually happens when the voice is installed. If a voice is missing, the README says to check the Windows TTS settings.

## Sources

- [EDCD/EDDI on GitHub](https://github.com/EDCD/EDDI)
- [Issues](https://github.com/EDCD/EDDI/issues)
- [README](https://github.com/EDCD/EDDI/blob/develop/README.md)
- [Releases](https://github.com/EDCD/EDDI/releases)

---

Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/edcd-eddi