Processing methods for the detection of landmark acoustic cues

Landmarks (Stevens, 2002) are acoustic cues that correlated with certain changes in speech articulation, and can be used to infer some of the distinctive features useful for recognition, largely manner features. This project identifies organizes processing steps involved extracting eight types Landmark a feature-based hierarchical automatic recognition system, which each module detects one landmark cue from samples continuous speech, based on knowledge about human production measurements such as formant frequency energy. It maps out necessary accurately detect Landmarks, describes standardized system modules implemented Matlab identify files quickly efficiently. The methods laid this provide framework later extended find other related place voicing.