[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:01317] Dependencies in HHEd context clustering



    Hello HTS-users;

In the HTS-SLT demo, during the HHEd context-clustering phases
(CXCL1 and CXCL2), the .hed files for cmp and lf0 call
the clustering commands according to the model's state order
(i.e., for 3 states models, first operate the clustering for
state 2, then for state 3, then for state 4 across the models).

In the way HHEd operates, does the overall result of the clustering
depend on the order of these commands?
More specifically:
does each TB command start from the originally loaded models,
hence all starting from the same MDL criterion value, or do they
start from the clustered models that resulted from the preceding
TB command, i.e., from the MDL criterion value reached after
the preceding TB command?
In the second case, has anyone tried to change the order of
the TB commands, and if yes to which extent did that impact
the synthesis results?

Same question with the dependencies of stream-related clustering:
does it matter to context-cluster first mgc, then lf0, then durations,
or can these clustering operations be safely considered orthogonal
from the point of view of MDL optimization?

Thanks in advance;
                              -*- Sacha K. -*-
--
Dr.Sacha Krstulovic - Research Engineer
Toshiba Research Europe Limited
Cambridge Research Laboratory
Speech Technology Group
208 Science Park, Milton Road
Cambridge CB4 0GZ - United Kingdom
Tel:    +44 1223 436 978
Fax:    +44 1223 436 909
E-mail: sacha@xxxxxxxxxxxxxxxxx



______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email ______________________________________________________________________
begin:vcard
fn:Sacha Krstulovic
n:Krstulovic;Sacha
org:Toshiba Research Europe Limited - Cambridge Research Laboratory;Speech Technology Group
adr:;;260 Science Park, Milton Road;Cambridge;;CB4 0WE;United Kingdom
email;internet:sacha@xxxxxxxxxxxxxxxxx
title:Research Engineer
tel;work:+44 1223 436 978
tel;fax:+44 1223 436 909
x-mozilla-html:FALSE
url:http://www.toshiba-europe.com/research/crl/
version:2.1
end:vcard


Follow-Ups
[hts-users:01318] Re: Dependencies in HHEd context clustering, Simon King
[hts-users:01319] Re: Dependencies in HHEd context clustering, Heiga ZEN (Byung Ha CHUN)