12.07.2015 Views

The HTK Book Steve Young Gunnar Evermann Dan Kershaw ...

The HTK Book Steve Young Gunnar Evermann Dan Kershaw ...

The HTK Book Steve Young Gunnar Evermann Dan Kershaw ...

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

3.1 Data Preparation 24onetwodialthree... etcsent-startzerosent-endJulianOdellphonecallDave... etc<strong>Steve</strong>Ollason<strong>Young</strong>Fig. 3.1Grammar for Voice DiallingGrammar(gram)HPARSEWord Net(w d n e t )Fig. 3.2Step 1<strong>The</strong> above high level representation of a task grammar is provided for user convenience. <strong>The</strong><strong>HTK</strong> recogniser actually requires a word network to be defined using a low level notation called<strong>HTK</strong> Standard Lattice Format (SLF) in which each word instance and each word-to-word transitionis listed explicitly. This word network can be created automatically from the grammar above usingthe HParse tool, thus assuming that the file gram contains the above grammar, executingHParse gram wdnetwill create an equivalent word network in the file wdnet (see Fig 3.2).3.1.2 Step 2 - the Dictionary<strong>The</strong> first step in building a dictionary is to create a sorted list of the required words. In the telephonedialling task pursued here, it is quite easy to create a list of required words by hand. However, ifthe task were more complex, it would be necessary to build a word list from the sample sentencespresent in the training data. Furthermore, to build robust acoustic models, it is necessary to trainthem on a large set of sentences containing many words and preferably phonetically balanced. Forthese reasons, the training data will consist of English sentences unrelated to the phone recognitiontask. Below, a short example of creating a word list from sentence prompts will be given. As notedabove the training sentences given here are extracted from some prompts used with the TIMITdatabase and for convenience reasons they have been renumbered. For example, the first few itemsmight be as follows

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!