Language Model, Acoustic Model, Pronunciation Model

[Isobel W]

I am getting really confused over the differences between the language model, the acoustic model and the pronunciation model.
I know the acoustic model is what the HMM calculates- the likelihood of a particular sequence of words for a model. The pronunciation model links things in the acoustic model to things in the language model (I am unsure how) . The language model I am most confused by because when I google it, all links are about n-grams and I don’t understand how n-grams fit into our model as I have never learnt about them before.

[Simon]

The language model emits a sequence of words.

The pronunciation model emits a sequence of phonemes, given a word.

The acoustic model of a phoneme emits a sequence of observations (e.g., MFCC feature vectors).

In order to compile these together into a recognition graph, all of them must be finite state.

The most common form of language model is an N-gram. This is a finite state model that emits a word sequence. The probability of a given word sequence is the product of the probability of each word in the sequence. Those word probabilities are computed given only the N-1 preceding words.

Another form of finite state language model is the hand-crafted grammar used in the digit recogniser exercise.

[Author]

Posts