[hts-users:02794] Re: Utterance generation

[Simon King <Simon.King@xxxxxxxx>]

On 20 Apr 2011, at 12:07, Xavi Gonzalvo wrote:

> 
> For example, first file in the demo data looks like this:
> 
> 73 id _71 ; name H* ; 
> 74 id _72 ; name H* ; 
> 75 id _73 ; name L-L% ; 
> 76 id _74 ; name H* ; 
> 77 id _75 ; name H* ; 
> 78 id _76 ; name L+H* ; 
> 79 id _77 ; name L+H* ; 
> 80 id _78 ; name L-L% ;
> 
> whereas I obtain this:
> 
> 71 id _71 ; name L-L% ; 
> 72 id _72 ; name L-L% ; 
> 73 id _73 ; name H* ; 
> 74 id _74 ; name H* ; 
> 75 id _75 ; name H* ; 
> 76 id _76 ; name H* ; 
> 77 id _77 ; name L+H* ; 
> 78 id _78 ; name L+H* ; 

One is just a permutation of the other. The order is not necessarily meaningful in Festival utterance files - you need to check which syllables these ToBI symbols are attached to - you may find you have identical utterance structures which (for reasons I can't explain) have a different numbering of the items.

Or, try comparing the HTS labels you get from your .utt file with the HTS labels you get with the demo.

Simon


-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.