RG 3-01 'Multilingual Speech Recognition'

Breaking the Unwritten Language Barrier (BULB)

Logo des Projektes BULB

The BULB project aims at supporting the documentation of unwritten languages with the help of automatic speech and language processing, in particular automatic speech recognition (ASR) and machine translation (MT). We will address the documentation of three mostly unwritten African languages of the Bantu family (Basaa, Myene and Embosi). 

Web site: http://www.bulb-project.org


The goal of SecondHands is to design a robot that can offer help to a maintenance technician in a pro-active manner. We see this robot as a second pair of hands that can assist the technician when he/she is in need of help. The project revolves around four fundamental concepts:

  • the design of a new robotic assistant;
  • a knowledge base to facilitate proactive help;
  • a high degree of human-robot interaction;
  • advanced perception skills to function in a highly dynamic industrial environment.

Web site: https://secondhands.eu

Automatische Rechtschreibanalyse

Logo des Projekts Automatische Rechtschreibanalyse

This project develops software for the diagnosis of orthographic skills in freely written texts from school students. For this researchers from the fields of language pedagogics, computer linguistics and natural language processing cooperate. It is the goald to create a tool that allows the analysis of freely written texts with respect to orthographic errors at low costs and efforts. This allows for the creation of indiviudal error profiles as well as large scale studies on the topic.

Web site: http://www.automatische-rechtschreibanalyse.de


The consortium RADICAL works on the rapid development and deployment of keyword search systems in new, previously unseen languages. The work is carried out within the project Babel and is funded by the Intelligence Advanced Research Projects Activity (IARPA)

State-of-the-art key word search systems work on the output of large vocabulary continous speech recognition systems in the form of word graphs and word-based posterior probabilities. Within the team RADICAL we work on the study of grapheme based ASR systems.