KIT-Bibliothek

18: Kognitive Systeme, Vorlesung, SS 2018, 02.07.2018

This audio or video file is copyrighted. Access is only allowed via computers of the Karlsruhe Institute of Technology (KIT).

Author

Alexander Waibel

Editor

KIT | Webcast

Participating institute

Institut für Anthropomatik und Robotik (IAR)

Genre

Vorlesung

Description

  • 0:00:00 Start
  • 0:00:05 Speech Recognition (Components)
  • 0:06:07 Language Models
  • 0:08:06 Language Model Performance
  • 0:10:24 Estimation of Language model Quality
  • 0:19:23 Decoder
  • 0:19:55 Decoder - Assembling the Pieces
  • 0:20:59 Decoding with Beam Search
  • 0:21:22 Beam vs. WER
  • 0:21:56 Improving Speed on Cooperative speech
  • 0:22:11 Viterbi Alignment
  • 0:22:59 Measuring Recognizer Performance
  • 0:23:22 Sloppy Speech
  • 0:24:17 Neural Nets
  • 0:25:36 Neural Language Model
  • 0:26:35 How good does it have to be
  • 0:30:48 Neural Language Processing and Machine Translation
  • 0:31:30 Natural Language Processing
  • 0:33:39 Speech Deployment
  • 0:37:56 Voice Agents
  • 0:43:49 Emotional Speech
  • 0:45:25 NLU with slot filling
  • 0:46:22 Intelligent System and a Language Transparent World
  • 0:47:36 How to pick a Research Project
  • 0:50:55 Human Interaction
  • 0:51:00 Connecting a Multilingual World
  • 0:52:08 Everyone Speaks English
  • 0:52:31 Human effort
  • 0:52:41 Laguage Transparence
  • 0:52:47 Language Text
  • 0:52:52 Social Text
  • 0:53:17 Lectures
  • 0:53:35 Human Language Challenges
  • 0:54:10 Body Language and Facial Expressions
  • 0:54:45 Can Technology provide a Solution
  • 0:54:59 Language is Ambiguous
  • 0:57:01 Neural Network
  • 0:58:10 In Image processing
  • 0:58:35 Exponential Increase in Computing
  • 0:59:03 English Text Copora
  • 0:59:36 Deep Neural Nets
  • 1:00:00 Conversational Speech
  • 1:00:23 Machine Translation
  • 1:01:05 Statistical Machine Translation
  • 1:01:33 Alignmenr/Decoding in MT
  • 1:01:54 Recurrent Neurak Nets
  • 1:02:14 RNN Encoder - Decoder
  • 1:03:09 MT Benchmarks - KIT Performance
  • 1:03:45 Consecutive Interpretation
  • 1:04:16 Interprating Machine
  • 1:05:49 First Speech Translation Videocall
  • 1:06:52 Jibbigo on Apple Commercials
  • 1:10:28 unlimited Domain Simultaneous
  • 1:14:49 Human-Machine symbiosis
  • 1:15:02 Voting Sessions
  • 1:16:07 German Compounding
  • 1:16:29 Words
  • 1:19:13 The Long Tail of Language
  • 1:19:44 Language Adaptive Networks
  • 1:21:39 Neural ''Interlingua''?
  • 1:22:56 Multi-Modal Translation
  • 1:23:46 Meeting of the Future
  • 1:23:59 Grand Challenges
  • 1:28:26 Neuronale Netze zur Drohnensteuerung
  • 1:28:38 Bitvraze Crazyflie
  • 1:29:08 Steuerung der Crazyflie
  • 1:30:39 Regelschleife
  • 1:31:49 Regelung mit Neuralen Netzen

Duration (hh:mm:ss)

01:33:48

Series

Kognitive Systeme, Vorlesung, SS 2018

Published on

05.07.2018

Subject area

Computer science

License

KITopen Licence

Resolution 1280 x 720 Pixel
Aspect ratio 16:9
Audio bitrate 128000 bps
Audio channels 2
Audio Codec aac
Audio Sample Rate 48000 Hz
Total Bitrate 934269 bps
Color Space yuv420p
Container mov,mp4,m4a,3gp,3g2,mj2
Media Type video/mp4
Duration 5628 s
Filename DIVA-2018-516_hd.mp4
File Size 657.256.581 byte
Frame Rate 25
Video Bitrate 800167 bps
Video Codec h264

Embed Code