Das LMU Logo

Institute of Phonetics and Speech Processing


WikiSpeech 2.22.0

Welcome to IPS WikiSpeech!

What's it all about?

WikiSpeech is a content management system for the web-based creation of speech databases for the development of spoken language technology and basic research. Its main features are support for the typical recording and project administration workflow, editing of the speech content and data download for the project administrator. WikiSpeech currently has been localized to English and German.

What do you have to do?

The project in which you participate. You will for example read simple texts: single words, commands, dates and times, phonetically balanced sentences as well as names and numbers. Besides you may answer a few questions and/or describe a few pictures.

What are we doing with your recordings?

The speech data will be transmitted to the IPS where they will be processed, documented and published. The sessions of most projects will be made anonymous. These projects use only anonymous personal properties such as your gender or age for publishing.

Projects (in alphabetical order):


    British Antarctic Survey Recordings


    Recordings in Argentina in collaboration with Hansjörg Mixdorff

  • Australian Voices
  • BA_BgDialekt

    Money terms in Ewe in Togo

  • BoDyS

    L2 learners of English in France, Université Paris Didérot




    Czech recordings 2009

  • Demo

    Demo project for WikiSpeech presentations and tests

  • Demo_mono
  • Demo_SSBE
  • Demo_stereo
  • DE_TR

    Estnische Aufnahmen von Mareike Plüschke

  • EnglishIntonation

    American and Jamaican English Intonation

  • Grundig-GBS_SPEAKER
  • HighVariabilityTraining
  • ICE-NG

    ICE Corpus Nigerian English