Subject: Speech Processing

Topic: Speech Processing Applications

Difficulty: Low

(i) Number of templates are limited.

(ii) It is only specific to a particular speaker.

(iii) Need actual training examples.

(iv) It can produce pathological results. The crucial observation is that the algorithm may try to explain variability in the Y-axis by warping the X-axis. This can lead to unintuitive alignments where a single point on one time series maps onto a large subsection of another time series.

(v) They suffer from the drawback that they may prevent the "correct" warping from being found. In simulated cases, the correct warping can be known by warping a time series and attempting to remove the original.

(vi) An additional problem with DTW is that the algorithm may fail to find obvious natural alignments in two sequences simply because a feature (i.e. peak, valley, inflection point, plateau etc.) in one sequence is slightly higher or lower that is corresponding feature in other sequence.

(vii) The weakness of DTW is in the features it considers.

(viii) The other main drawback of DTW is the explosion of the search space for continuous recognition tasks and poor speaker independent performance.

(ix) One of the main problems in dynamic time warping based speech recognition systems is the preparation of reliable reference templates for the set of words to be recognized.