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

[hts-users:02461] Re: problem building regression-class trees and adaptation problem with xform mask

Subject: [hts-users:02461] Re: problem building regression-class trees and adaptation problem with xform mask
From: Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
Date: Thu, 6 May 2010 14:46:13 +0900
Cc: uratec <uratec@xxxxxxxxxxxxxxx>
Delivered-to: hts-users@xxxxxxxxxxxxxxx
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:sender:received :in-reply-to:references:date:x-google-sender-auth:message-id:subject :from:to:cc:content-type:content-transfer-encoding; bh=YDs1UPJ9IMF+h9q5HXY7AM114jcW9OGPq3lPktrVsvw=; b=qPmMF3/PPRsiZDbLeiKrJNZianUkyIGuJoACNMCJOcS3pls8lFEHj0J4RnVAV8BcTp EFVVqE2W/h7j4fsRyjrxEKkB9DLkfmVc2tctgs0WVpeIyCJOLD8eJB9VJhFWgq1aBCNf FptrbNJ0iRmQgn8OeziehJP9y0G4q35yuPk4A=
Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=anLlbdieycEhl2SAlNrQPgz7NYNo9m5pRsk8fSZD2r3MH7EGIiWsdO3xzeq7F+zf+l b/JiWOyO1FxJ5rYy23GXKlQS4knVMWvUi+/m7yCkhKHqacuDPmJFbi1mvPwINeTVpGUE H3KLPZ3PT0NQCVHgT+UKtDhmMu69dUTRMVvEQ=

Hi,

First error:
How large database do you use?
Does your machine have enough memory?

Second error:
Please, set appropriate speaker name pattern to -h option.
If filenames are Mike_001.cmp and John_001.cmp, */%%%%_* should be set.

Regards,
Keiichiro Oura

2010/5/5 Alexandros Lazaridis <alaza@xxxxxxxxxxxxxxxxx>:
> Hello,
>
>
>
> I have tried the HTS-demo_CMU-ARCTIC-ADAPT and the HTS-demo_CMU-ARCTIC-SLT
> and are working fine.
>
> I tried the adaptation using my one database and I am getting the following
> error.
>
>
>
> FIRST ERROR
>
>
>
> ========================================================================================
>
> Start building regression-class trees for adaptation at Sun May  2 07:05:03
> GTBST 2010
>
> ========================================================================================
>
>
>
> /cygdrive/c/htk/bin/HHEd -A -B -C
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/trn.cnf -D -T 1 -p -i
> -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/re_clustered.mmf
> -M
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/edfiles/qst001/ver1/cmp/reg.hed
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/lists/full.list
>
>
>
> HTK Configuration Parameters[5]
>
>   Module/Tool     Parameter                  Value
>
> #                 VFLOORSCALESTR  Vector 4 0.01 0.01 0.01 0.01
>
> #                 MINLEAFOCC              5.000000
>
> #                 NATURALWRITEORDER              TRUE
>
> #                 NATURALREADORDER              TRUE
>
> #                 APPLYVFLOOR                 TRUE
>
>
>
> HHEd
>
>  48159/48159 Models Loaded [7 states max, 2 mixes max]
>
>
>
> // load stats file
>
> LS
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/stats/qst001/ver1/cmp.stats.re-clustered
>
>  Loading state occupation stats
>
>   Stats loaded for 48159 models
>
>   Mean Occupation Count = 1183.030462
>
>
>
> // construct regression class tree
>
> RC 32 reg
>
>  Building regression tree with 32 terminals (4 streams)
>
> Creating regression class tree with ident reg.tree and baseclass reg.base
>
> Splitting Node 1, score 1.000000e+10 (Stream splitting)
>
> Splitting Node 3, score 1.000000e+10 (Stream splitting)
>
> Splitting Node 5, score 1.000000e+10 (Stream splitting)
>
> Splitting Node 7, score 1.000000e+10 (MSD splitting)
>
> Splitting Node 6, score 1.000000e+10 (MSD splitting)
>
> Splitting Node 4, score 1.000000e+10 (MSD splitting)
>
> Splitting Node 2, score 2.342814e+06 (Stream=1, vSize=75)
>
> Splitting Node 14, score 8.727774e+05 (Stream=1, vSize=75)
>
> …………………..
>
> ………………….
>
> ………………………
>
> Splitting Node 162, score 2.965114e+01 (Stream=4, vSize=1)
>
> Splitting Node 169, score 2.741338e+01 (Stream=4, vSize=1)
>
> No more nodes to split..
>
> stream 1: #terminals=10
>
> stream 2: #terminals=32
>
> stream 3: #terminals=32
>
> stream 4: #terminals=32
>
> 2 [main] HHEd 3624 _cygtls::handle_exceptions: Error while dumping state
> (probably corrupted stack)
>
> Error in /cygdrive/c/htk/bin/HHEd   -A -B -C
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/trn.cnf -D -T 1 -p -i
> -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/re_clustered.mmf
> -M
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/edfiles/qst001/ver1/cmp/reg.hed
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/lists/full.list
>
>
>
> The thing is that when I reduced the files (raw and utt files) of the
> database I use, this error didn’t occur. As I can understand this error has
> to do with the memory. I kwon that I should avoid windows and cygwin but is
> there a way to overcome this error without having to reduce the size of the
> database I use?
>
>
>
>
>
> The second error was that when I used the reduced database I got another
> error at the stage of the “estimating transforms for speaker adaptive
> training (SAT) (1st iteration)”
>
>
>
> SECOND ERROR
>
> ===================================================================================================================
>
> Start estimating transforms for speaker adaptive training (SAT) (1st
> iteration) at Tue May  4 20:59:35 GTBST 2010
>
> ===================================================================================================================
>
>
>
> /cygdrive/c/htk/bin/HERest -A -B -C
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/trn.cnf -D -T 1 -S
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/scp/train.scp -I
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/labels/full.mlf -m 1 -u
> tmvw -w 3 -t 1000 100 5000 -h */rich_sentences_%%%_* -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/re_clustered_sat.mmf
> -M /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp -K
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/xforms
> dec_sat1 -u a -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.base
> -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.tree
> -C /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/sat.cnf
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/lists/full.list
>
>
>
> HTK Configuration Parameters[13]
>
>   Module/Tool     Parameter                  Value
>
> #                 BANDWIDTH       IntVec 3 24 24 24 IntVec 1 0 IntVec 1 0
> IntVec 1 0
>
> #                 BLOCKSIZE       IntVec 3 25 25 25 IntVec 1 1 IntVec 1 1
> IntVec 1 1
>
> #                 SPLITTHRESH     Vector 4 10000.0 2000.0 2000.0 2000.0
>
> #                 USEBIAS                     TRUE
>
> #                 TRANSKIND                  CMLLR
>
> #                 REGTREE
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.tree
>
> #                 BASECLASS
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.base
>
> #                 ADAPTKIND                   TREE
>
> #                 VFLOORSCALESTR  Vector 4 0.01 0.01 0.01 0.01
>
> #                 MINLEAFOCC              5.000000
>
> #                 NATURALWRITEORDER              TRUE
>
> #                 NATURALREADORDER              TRUE
>
> #                 APPLYVFLOOR                 TRUE
>
>
>
> HERest  ML Updating: XForms
>
>
>
>  System is SHARED
>
> 21320 Logical/21320 Physical Models Loaded, VecSize=78
>
> 3 MMF input files
>
> Pruning-On[1000.0 100.0 5000.0]
>
>   ERROR [+999]  Output xform mask */rich_sentences_%%%_* does not match
> filename
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/cmp/orientel/rich_sentences_orientel_0041.cmp
>
>  FATAL ERROR - Terminating program /cygdrive/c/htk/bin/HERest
>
> Error in /cygdrive/c/htk/bin/HERest -A -B -C
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/trn.cnf -D -T 1 -S
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/scp/train.scp -I
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/labels/full.mlf -m 1 -u
> tmvw -w 3 -t 1000 100 5000 -h "*/rich_sentences_%%%_*" -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/re_clustered_sat.mmf
> -M /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp -K
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/xforms
> dec_sat1 -u a -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.base
> -H
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/models/qst001/ver1/cmp/regTrees/dec.tree
> -C /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/configs/sat.cnf
> /cygdrive/c/htk/HTS-demo_WCL-GREEK-MARIA-ADAPT/data/lists/full.list
>
>
>
> Thank you in advance,
>
> Alexandros Lazaridis
>
>
>
> ========================================
>
> Alexandros Lazaridis
>
> Electrical and Computer Engineer
>
> PhD Student
>
>
>
> Artificial Intelligence Group
>
> Wire Communications Laboratory
>
> Department of Electrical & Computer Engineering
>
> University of Patras
>
> Patras 265 00, Greece
>
>
>
> Email: alaza@xxxxxxxxxx
>
> Tel:  +30 2610 996496
>
> Fax: +30 2610 997336
>
> http://www.wcl.ece.upatras.gr/ai/
>
> http://www.wcl.ece.upatras.gr/alaza/
>
> BRING THEM BACK
>
> http://www.bringthemback.org/
>
> ========================================
>
>
>
>

Follow-Ups
: [hts-users:02462] Re: problem building regression-class trees and adaptation problem with xform mask, Alexandros Lazaridis

References
: [hts-users:02460] problem building regression-class trees and adaptation problem with xform mask, Alexandros Lazaridis

Prev by Subject: [hts-users:02460] problem building regression-class trees and adaptation problem with xform mask
Next by Subject: [hts-users:02462] Re: problem building regression-class trees and adaptation problem with xform mask
Previous by thread: [hts-users:02460] problem building regression-class trees and adaptation problem with xform mask
Next by thread: [hts-users:02462] Re: problem building regression-class trees and adaptation problem with xform mask