en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Open Vocabulary Speech Analysis in VITALAS
Published on Feb 18, 20086431 Views
Automatic indexing of TV and radio speech data requires robust components for both speech recognition and spoken document retrieval. Due to the high topic variability and the resulting large vocabular
Related categories
Chapter list
Open Vocabulary Speech Analysis in Vitalas00:00
Outline00:25
Challenge in Vitalas: Large Scale Broadcast News Indexing01:02
Structural Audio Analysis in Vitalas01:54
Speech Recognition02:33
Speech Recognition Challenges03:26
Phonetic Approach to Open Vocabulary Indexing - 105:04
Phonetic Approach to Open Vocabulary Indexing - 205:24
Phonetic Approach (1): Generate Subword Transcription05:57
Phonetic Approach (2): Fuzzy Syllable Search06:24
Properties of Phonetic Subword Approach07:32
Experiments: Fraunhofer AudioMining Corpus09:31
Experiments: Model Setup11:24
Current Results - Speech Recognition12:09
Current Results – Fuzzy Phonetic Retrieval13:18
Additional Word Context for Enhanced Display of Results14:47
Demo: AudioMining15:29
Next Steps19:38
Thank you!20:21