Home immigration Germany to use SPEECH RECOGNITION to pin down countries of origin of...

Germany to use SPEECH RECOGNITION to pin down countries of origin of thousands of refugees


Germany is bringing in state-of-the-art speech recognition technology to pin down the countries of origin of thousands of refugees.

The Federal Office for Migration and Refugees (BAMF) will begin testing a computer programme dubbed ‘the tongue’ within a fortnight.

Join today
Get your digital copy today!

Get your digital copy today!

Sign up to receive the latest news and updates from Knights Templar International and get regular copies of our publications sent straight to your inbox!

Thank you, you will recieve your digital copy shortly!  Please check your spam/junk email to confirm.  We do not want you to miss out on this.

Computer software has been developed to analyse and identify dialects of people seeking asylum in Germany, based on recorded speech samples.

That information could then be used to pin down exactly where individuals hail from.

The technology is based on voice authentication software used by banks and insurance companies which have been modified to analyse dialects.

Some 60 per cent of migrants who arrived in Germany in 2016 did not possess papers.

In a bid to speed up the identification of all migrants a new law was passed this year authorising officials to probe laptops and mobile phones in their efforts to ascertain asylum seekers’ identities.

Monika Schmid, a linguistics professor at the University of Essex, said that speech analysts “must have a solid background in linguistic analysis and be able to take into account a wide range of factors” when determining someone’s home country.

She voiced concerns over the new programme, stating: “I don’t see how automated software can distinguish whether a person uses a certain word or pronounces it in a particular way because this is part of their own repertoire or because they were primed to do so by the interviewer or interpreter.”

Dirk Hovy, a computer scientist at the University of Copenhagen, said that BAMF’s system would need to incorporate speech data that is demographically representative of asylum seekers, which would involve the creation of a broad database.

He said: “Creating a perfect dataset is virtually impossible because language is constantly changing.”

Join today