N-gram model

Galescu, L., and Allen, J. Name Pronunciation with a Joint N-Gram Model for Bidirectional Grapheme-to-Phoneme Conversion. Proc. ICSLP (2002), 109-112. 

The idea is to measure the probability of this for the sequence of frames across every pair of candidate units across a join, and use the result as a measure of join qnality in the search. P 0) is too difiicnlt to estimate in full, so we make the n-gram assumption and estimate it on a shorter seqnence of frames (say two before and two after the join). The powerful thing abont this approach is that the model can be trained on all the data in the speech database, not jnst examples near unit boundaries. This often results in many... 

Texture measurements Texture of canned carrots was measured using Instron Universal Testing Machine (Model 1011) fitted with Kramer shear cell. Thirty grams of drained carrot cubes were evenly placed in the Kramer shear cell and were compressed, sheared and extruded using a crosshead speed of 100 mm/min. Each measurement was repeated 10 times and the mean was used to express the firmness of carrot cubes in Newton(N). 

---