[hts-users:02603] Component specified multiple times
- Subject: [hts-users:02603] Component specified multiple times
- From: ali azimizadeh <ali.azimizadeh@xxxxxxxxx>
- Date: Tue, 31 Aug 2010 15:08:06 +0330
- 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:received:date:message-id :subject:from:to:content-type; bh=D+QOd99BwzFDFHWMOM1BC8G6JSziEeO+9Q2X5iUFYOo=; b=vyRwa/K1B74gNsQcb/rw3BmYrY+OA5G3mL36J52Jx811uP6vJkW9mh8Ku8OOoEvI8K v5s7+JOXgguMNp3LqZlNCO0T7SjvXbKww2XnTD9Lbggp//SNBcRMR8L90p129QXUMb38 Jcst2GLgpYHw5KkJkAy+pp7gCUUZZv0S0j8XM=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=qfpZUfuapudA2uG9os/ylW+gzDXb6x+s2xb8KvWzy+w7GQ6EII8SLA6ll+5kyR4YQv dDdx6we5xA3PMgbJyPpWpq46VIJdmXayO9ZcQLnPEXNRjIQ3P/TvYs+8FUra27+s95Fb /0vysSD+roLpqnnVZJ8QEl3145r9rKzih+YRA=
Hi,
I am going to implement HTS-ADAPT 2.1.1Beta by my own dataset. i have just one speaker. i set 100 utterances of training data as adaptation data. finally, i give this error:
Splitting Node 247, score 3.524929e+01
(Stream=4, vSize=1)
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm r^a$-n+e=s$@1_2/A:0_0_2/B:0-0-2@1-2&14-5#4-3$3-2!2-1;2-4|e/C:1+0+3/D:P_1/E:N_SING+2@6+2&6+1#1+1/F:V_PRS_3/G:7_3/H:18=7@2=2|L-H%/I:17=8/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm a^m-d+i=d@1_3/A:0_0_3/B:0-0-3@1-3&3-15#1-6$1-4!2-1;0-2|i/C:1+0+2/D:CON_1/E:N_SING+3@3+6&3+5#1+1/F:N_SING_3/G:18_7/H:17=8@3=1|L-L%/I:0=0/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm h^a-m+n=e@3_1/A:0_1_1/B:0-0-3@1-1&1-18#1-6$1-4!0-3;0-4|a/C:0+0+2/D:ADV_NI_4/E:CON+1@1+7&1+6#0+1/F:N_SING_4/G:7_3/H:18=7@2=2|L-H%/I:17=8/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm h^a-m+n=e@3_1/A:0_1_1/B:0-0-3@1-1&1-18#1-6$1-4!0-3;0-4|a/C:0+0+2/D:ADV_NI_4/E:CON+1@1+7&1+6#0+1/F:N_SING_4/G:7_3/H:18=7@2=2|L-H%/I:17=8/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm h^a$-i+#=h@1_1/A:1_0_2/B:0-1-1@4-1&7-1#3-1$1-1!1-0;0-0|i/C:0+0+3/D:N_SING_2/E:ADV_NI+4@3+1&3+0#1+0/F:CON_1/G:0_0/H:7=3@1=3|H-H%/I:18=7/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm z$^e-b+e=t@1_2/A:1_0_2/B:0-0-2@1-4&4-4#2-2$1-2!1-2;0-3|e/C:0+0+3/D:N_SING_2/E:ADV_NI+4@3+1&3+0#1+0/F:CON_1/G:0_0/H:7=3@1=3|H-H%/I:18=7/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm r^s-a$+r=a$@2_1/A:0_0_3/B:1-1-2@2-1&12-7#3-3$2-2!3-3;7-6|a$/C:0+0+2/D:N_PL_5/E:ADJ_SIM+2@4+4&4+3#1+1/F:P_1/G:7_3/H:18=7@2=2|L-H%/I:17=8/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
WARNING [-2671] CalcClusterDistribution: Accumulates are non existant for hmm e^p-a$+r=s@2_2/A:0_0_2/B:0-0-3@1-2&11-8#3-4$2-3!2-1;6-1|a$/C:1+1+2/D:N_PL_5/E:ADJ_SIM+2@4+4&4+3#1+1/F:P_1/G:7_3/H:18=7@2=2|L-H%/I:17=8/J:42+18-3 in /root/Desktop/festival/htk/bin/HHEd
Splitting Node 250, score 3.449984e+01
(Stream=4, vSize=1)
No more nodes to split..
stream 1: #terminals=11
stream 2: #terminals=32
stream 3: #terminals=32
stream 4: #terminals=32
stream 5: #terminals=32
ERROR [+999] CheckBaseClass: Component specified multiple times
ERROR [+999] BaseClass check failed
FATAL ERROR - Terminating program /root/Desktop/festival/htk/bin/HHEd
Error in /root/Desktop/festival/htk/bin/HHEd -A -B -C /root/Desktop/festival/hts_demo_adapt_beta/configs/trn.cnf -D -T 1 -p -i -H /root/Desktop/festival/hts_demo_adapt_beta/models/qst001/ver1/cmp/re_clustered.mmf -M /root/Desktop/festival/hts_demo_adapt_beta/models/qst001/ver1/cmp/regTrees /root/Desktop/festival/hts_demo_adapt_beta/edfiles/qst001/ver1/cmp/reg.hed /root/Desktop/festival/hts_demo_adapt_beta/data/lists/full.list
after checking cmp.stats.re-clustered, reg.tree and reg.base and c codes, i concluded that there is a class in reg.base that had repeated 33 times.
<CLASS> 5 {n^d-#+x=x@x_x/A:0_1_3/B:x-x-x@x-x&x-x#x-x$x-x!x-x;x-x|x/C:0+0+0/D:V_PA_3/E:x+x@x+x&x+x#x+x/F:0_0/G:7_3/H:x=x@1=5|0/I:0=0/J:38+19-5.state[2].stream[3].mix[1],a^r-#+x=x@x_x/A:0_1_3/B:x-x-x@x-x&x-x#x-x$x-x!x-x;x-x|x/C:0+0+0/D:CON_5/E:x+x@x+x&x+x#x+x/F:0_0/G:6_2/H:x=x@1=1|0/I:0=0/J:6+2-1.state[2].stream[3].mix[1]}
<CLASS> 6 {n^d-#+x=x@x_x/A:0_1_3/B:x-x-x@x-x&x-x#x-x$x-x!x-x;x-x|x/C:0+0+0/D:V_PA_3/E:x+x@x+x&x+x#x+x/F:0_0/G:7_3/H:x=x@1=5|0/I:0=0/J:38+19-5.state[2].stream[3].mix[1],a^r-#+x=x@x_x/A:0_1_3/B:x-x-x@x-x&x-x#x-x$x-x!x-x;x-x|x/C:0+0+0/D:CON_5/E:x+x@x+x&x+x#x+x/F:0_0/G:6_2/H:x=x@1=1|0/I:0=0/J:6+2-1.state[2].stream[3].mix[1]}
...
when i search this label in cmp.stats.re-clustered there was just one label without any repeats. also, i can't see any special characteristic in that. can any one tell me how reg.base file is created?
also, i can remove duplicates from reg.base but the reg.tree won't match with that, can i change the reg.tree to match with new reg.base?
Thanks a lot for your attention
-Ali