07: Kognitive Systeme, Vorlesung, SS 2017, 31.05.2017
- 0:00:00 Starten
- 0:00:10 Lecture Demos
- 0:00:42 Spectogram Controls
- 0:09:49 Speech: State-of-the-Art
- 0:14:09 Sloppy Speech
- 0:16:47 Speech Recognition (System Overview)
- 0:20:21 Voiced and Unvoiced Phonemes
- 0:21:35 Analog to Digital
- 0:21:50 Front End Processing
- 0:23:49 Linear Sequence Alignment
- 0:25:15 Problem with Linear Alignment
- 0:28:08 Speech Recognition (Components)
- 0:30:50 Spectogram
- 0:32:04 Markov Models
- 0:36:18 Single Fair Coin
- 0:37:23 Discrete Observation HMM
- 0:40:54 Hidden Markov Models
- 0:45:25 Acoustic Modeling
- 0:48:13 HMM Problems and Solutions
- 0:51:35 Evaluation
- 0:54:08 The Forward Algorithm
- 0:58:58 Forward Trellis
- 1:04:39 The Backward Algorithm
- 1:05:00 Decoding
- 1:08:35 Viterbi Trellis
| Zugehörige Institution(en) am KIT | Institut für Anthropomatik und Robotik (IAR) |
| Publikationstyp | Audio & Video |
| Publikationsdatum | 09.06.2017 |
| Erstellungsdatum | 31.05.2017 |
| Sprache | Deutsch |
| DOI | 10.5445/DIVA/2017-319 |
| Identifikator | KITopen-ID: 1000114719 |
| Lizenz | KITopen-Lizenz |
| Serie | Kognitive Systeme, Vorlesung, SS 2017 |
| Folge | 7 |