MLLP members travelled to the European Parliament in Brussels on 4 September to make a presentation on our most recent technology for multilingual real-time speech-to-text. See here our demo videos.
The MLLP has been polishing in the last few months a new application of its self-developed technologies: real-time speech-to-text transcription with automatic language detection. An example of a use case for this technology would be the transcription in real-time of European Parliament sessions, with each participant speaking in their own language. So far, we have added support for real-time speech-to-text with the automatic detection of 7 languages: English, German, French, Italian, Spanish, Dutch and Portuguese.
On 4 September, MLLP members Alfons Juan and Joan Albert Silvestre were invited by the European Parliament (EP) to its premises in Brussels for a presentation on the current state of this technology. Attending the meeting were several members of the EP’s Directorate-General for Interpretation and Conferences (DG INTE).
Alfons and Joan Albert presented the state of the technology, including two demo videos of the technology in use. Then, the attendants to the meeting were able to try themselves the technology in any of the supported languages through an on-line service hosted at the MLLP servers at Universitat Politècnica de València.
We invite you to watch the two demo videos that were shown in the meeting and see for yourself:
Attendants to the meeting stated their surprise at the high accuracy and fast response of the service. We at the MLLP would like to thank the EP for this occasion to share our experience with this technology. We think now is the right time to increase the use of language technologies as support tools to boost productivity.