transLectures @Internet of Education 2013

311
-1

Published on

An overview of transLectures, plus results at the end of the second year of the project.
Internet of Education 2013, 11-12 Nov 2013, Ljubljana, Slovenia.

Video of the presentation available online at: http://videolectures.net/internetofeducation2013_diaz_munio_translectures/

Published in: Business, Technology
1 Comment
0 Likes
Statistics
Notes
  • Watch the video of this presentation at: http://videolectures.net/internetofeducation2013_diaz_munio_translectures/
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

No Downloads
Views
Total Views
311
On Slideshare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
1
Comments
1
Likes
0
Embeds 0
No embeds

No notes for slide

transLectures @Internet of Education 2013

  1. 1. Internet of Education 2013 12 November 2013 Gonçal Garcés Díaz-Munío Universitat Politècnica de València EC FP7 ICT project #287755
  2. 2. Outline • transLectures (tL): motivation and goals • Video demo • tL technologies • Progress and results • Scientific evaluations • User evaluations • Quality control (expert evaluations) • Implementation and integration • tL open source tools 12 Nov 2013 2
  3. 3. Motivation • Video lecture repositories and MOOCs • Thousands of hours of video lectures available • Hundreds of hours of video lectures recorded every week • Most video lectures only available in their original language • No subtitles 12 Nov 2013 3
  4. 4. Motivation • Transcriptions and translations are needed • • • • • Accessibility for people with disabilities Accessibility for speakers of different languages Search and analysis functions Automated topic finding … 12 Nov 2013 4
  5. 5. Motivation • Transcriptions and translations are needed • • • • • Accessibility for people with disabilities Accessibility for speakers of different languages Search and analysis functions Automated topic finding … • How do we get there? 12 Nov 2013 5
  6. 6. The transLectures approach 1. Automatic Speech Recognition (ASR) and Machine Translation (MT) • Adaptation: Taking advantage of the characteristics of video lecture repositories • High-quality automatic transcriptions and translations 2. Interactive postediting: intelligent interaction for reduced effort 12 Nov 2013 6
  7. 7. Goals • Massive adaptation • Intelligent interaction • Implementation • Case studies: Videolectures.NET & Polimedia • Real-life evaluation • Integration into Opencast Matterhorn http://opencast.org/matterhorn/ 12 Nov 2013 7
  8. 8. The transLectures partners 1 2 Name Universitat Politècnica de València Xerox SAS Country Spain France 3 3+ 4 Institut Jožef Stefan Knowledge for All Foundation RWTH Aachen University Slovenia UK Germany 5 6 EML – European Media Laboratory DDS – Deluxe Digital Studios Germany UK 12 Nov 2013 8
  9. 9. Languages • Transcription (ASR) • EN • SL • ES • Translation (MT) • • • • EN>SL , SL>EN EN>ES , ES>EN EN>FR EN>DE 12 Nov 2013 9
  10. 10. transLectures: video demo http://www.translectures.eu/new-demo-video/ 12 Nov 2013 10
  11. 11. Massive adaptation • Characteristics of video lectures Just one person Known speaker Clear talking No interruptions Focused on a topic Slides 12 Nov 2013 11
  12. 12. Massive adaptation • Known speaker and topic • Slides • Related documents 12 Nov 2013 12
  13. 13. Scientific evaluations (Y2) • Transcription results Worse • WER: Word Error Rate (%) • Goal: WER < 25% • EN, SL, ES Better 12 Nov 2013 13
  14. 14. Scientific evaluations (Y2) • Translation results Better • BLEU • Goal: BLEU > 30 • • • • EN>SL , SL>EN EN>ES , ES>EN EN>FR EN>DE Worse 12 Nov 2013 14
  15. 15. Y1 results and comparison 12 Nov 2013 15
  16. 16. Y1 results and comparison 12 Nov 2013 16
  17. 17. Y1 results and comparison 12 Nov 2013 17
  18. 18. Intelligent interaction • Postediting automatic transcriptions/translations • The user invests the least possible effort • The system learns the most from it • Confidence measures • Fast constrained search 12 Nov 2013 18
  19. 19. Intelligent interaction 12 Nov 2013 19
  20. 20. Intelligent interaction 12 Nov 2013 20
  21. 21. Intelligent interaction • Batch interaction vs. Intelligent interaction 12 Nov 2013 21
  22. 22. Intelligent interaction • User evaluations • UPV (Polimedia) • JSI (Videolectures.NET) 12 Nov 2013 22
  23. 23. User evaluations • User evaluations at UPV • Users: lecturers • Revising their own lectures • 3 different experiments 1. Complete supervision 2. Intelligent interaction 3. Two-round supervision 12 Nov 2013 23
  24. 24. User evaluations 1. Complete supervision 12 Nov 2013 24
  25. 25. User evaluations 2. Intelligent interaction 12 Nov 2013 25
  26. 26. User evaluations 3. Two-round supervision 12 Nov 2013 26
  27. 27. User evaluations • User evaluations at UPV: results 12 Nov 2013 27
  28. 28. User evaluations • User evaluations at UPV: results 12 Nov 2013 28
  29. 29. User evaluations • User evaluations at UPV: results 12 Nov 2013 29
  30. 30. Quality control: expert evaluations • Transcription quality (EN, ES, SL) • UPV: Representative set of Spanish transcriptions • Avg. WER: 23.2 ; Avg. RTF: 3.8 12 Nov 2013 30
  31. 31. Quality control: expert evaluations • Transcription quality (EN, ES, SL) • UPV: Representative set of Spanish transcriptions • Avg. WER: 23.2 ; Avg. RTF: 3.8 • Translation quality (EN<>SL, EN<>ES, EN>FR, EN>DE) • UPV: Representative set of Spanish>English translations • Avg. BLEU: 46.6 ; Avg. RTF: 14.8 ; Avg. Score: 3.6 out of 5 12 Nov 2013 31
  32. 32. Implementation and integration • Videolectures.NET • Polimedia • Opencast Matterhorn 12 Nov 2013 32
  33. 33. • Polimedia 12 Nov 2013 33
  34. 34. transLectures: Open source tools • The tL player (& editor) • Coming soon (www.translectures.eu) • The transLectures-UPV Toolkit (TLK) for ASR • www.translectures.eu/tlk • RWTH Aachen: rASR, Jane (MT) • http://www-i6.informatik.rwth-aachen.de/web/Software/ 12 Nov 2013 34
  35. 35. Next steps for transLectures • Keep improving ASR and MT results • Keep improving tL open source tools (TLK, tL player) • External user evaluations (VL.NET and polimedia) • External trials: implementation in other universities 12 Nov 2013 35
  36. 36. www.translectures.eu • About this presentation: Gonçal Garcés • Project coordinator: Alfons Juan-Ciscar ggarces@dsic.upv.es ajuan@dsic.upv.es EC FP7 ICT Programme – Project Number 287755 12 Nov 2013 36

×