Automatic Syllable Detection for Vowel Landmarks

1. Introduction: LAFF and Vowel Landmarks
2. Background: Automatic Syllable Detection
3. Science: Studies of vowels in TIMIT database
   • Vowels in the TIMIT database
   • Experiment 1: Presence of F1 peak in vowels
   • Experiment 2: Coincidence of amplitude and frequency peaks in vowels
   • Experiment 3: Vowel quality at F1 peak vs. midpoint
   • Experiment 4: Fixed energy band vs. formant tracking
4. Engineering: Development of Vowel Landmark Detector
   • Scoring technique
   • Baseline experiment
   • Combining acoustic cues
   • Incorporating MLP into Detector
   • Evaluation and error analysis
   • Confidence scoring
5. Conclusion: Implications and Future Directions