You can either download one of the stable releases or the development version on the GitHub repository.

For all versions: Read the user guide for more instructions on how to use OpenDial.

OpenDial version 1.5 (released on 05.04.2016)

Bleeding-edge version:

If you prefer to work with the code on the GitHub repository, simply enter the following to checkout the code:

 git clone


A collection of plugins have been developed to extend OpenDial with new modules.

  • Nuance Speech: Speech recognition and synthesis using Nuance's cloud-based speech API.
  • AT&T Speech (Removed in the last version due to discontinued support, see AT&T website)
  • Mary TTS: Local speech synthesis using the open-source Mary TTS engine.
  • Sphinx ASR: Local speech recognition using the open-source Sphinx 4 ASR engine.
  • MaltParser: data-driven dependency parsing using the MaltParser and the Stanford POS tagger.

Starting from version 1.4 of OpenDial, all plugins are included by default in the release package, so you don't have to download anything. The external libraries for the Mary TTS, Sphinx ASR and MaltParser plugins are, however, only included in the "full package".

If you own a Nao robot and would like to implement some bindings between the NaoQi API and OpenDial, you can check this code.