high quality service affordable prices rapid project turnaround quotes sent within 1 hour
Phone: +44(0)118 958 4934

Matinee Blog

voice recognition software

Sloppy Software: When Voice Recognition Technology Doesn’t Quite Work

There is nothing like the natural flow of the human voice for making a point clearly, intuitively, and creatively. It’s very important, in contexts like video translation or voice over translations, to get good quality translation services so that all of the nuances of the source text can be captured.

Unfortunately, when people use cheap and ineffective voice recognition and voice translation software, terrible (and sometimes hilarious) things happen. This list of a few recent bloopers, broadcast to the world, will surely help you to think carefully about the need to get great quality IVR voice prompts and video translation services for your own business or personal needs

The call for ‘a moment’s violence’ (instead of a moment’s silence) at the Queen Mother’s funeral, and a reference to the Archbishop of Canterbury as the ‘Arch Bitch of Canterbury’ are just two of the high profile mistakes that were made in voice recognition and transcription in recent years.

As several articles, like this one from the Telegraph, on the topic showed, many viewers found these bloopers absolutely hilarious, whilst many others – among them a large group of Deaf viewers – were extremely frustrated.

The website ‘Page 888’ almost obsessively catalogues recent voice transcription and subtitling errors. Perhaps, unsurprisingly, this tends to happen most often with foreign or unfamiliar words, for instance the Welsh cheese Caerphilly being transcribed as ‘car Philippe Baku’.

So, when it comes to video translations from one language to another, it is extra important to ensure that top quality technology and personnel are used.

Watch any TV screen in a crowded and noisy pub, and they will often have automatically generated subtitles on for all the TV programmes they show. Why not watch them the next time you see them, and see if you can spot any particularly good mistakes which make you laugh, or just make you glad you’re more careful with linguistic technology.