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

[hts-users:03184] Re: Problem of building regression tree (average voice model)


Hi,

It is unlikely that having 20 speakers have caused this problem.  I expect that some data files are corrupted, i. e., failed to extract speech features from waveforms.  Please check whether your data is OK or not.

Regards,

Heiga

2012/02/24 9:23 "li jay" <lij.acd@xxxxxxxxx>:
Hi,

Thank you for telling me this. Compared to the delta f0 of the speaker dependent model I trained previously, it is too big ( the current model I'm training is an average voice model ) . I thought it was because I used speech sentences of 20 speakers, which resulted in big delta f0. I used the same configuration and option settings of speaker dependent model training and replaced the training data with the data of 20 speakers. Is it correct to train a average voice model without modify any configuration and option?

Regards,
Jay

2012/2/24 Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
Hi,

It seems that delta lf0 is not trained correctly.

<VARIANCE> 1
2.078870e+12

The variance is too big.
You should check delta f0 sequence in training data.

Regards,
Keiichiro Oura

2012/2/24 li jay <lij.acd@xxxxxxxxx>:
> Hi,
>
> The following is the part of re_clustered.mmf
>
>  170535 ~p "lf0_s2_523-3"
>  170536 <STREAM> 3
>  170537 <NUMMIXES> 2
>  170538 <MIXTURE> 1 1.582146e-01
>  170539 <MEAN> 1
>  170540  -9.624681e-03
>  170541 <VARIANCE> 1
>  170542  2.078870e+12
>  170543 <GCONST> 3.020072e+01
>  170544 <MIXTURE> 2 8.417839e-01
>  170545 <MEAN> 0
>  170546 <VARIANCE> 0
>  170547 <GCONST> 0.000000e+00
>  170548 ~p "lf0_s2_523-4"
>  170549 <STREAM> 4
>  170550 <NUMMIXES> 2
>  170551 <MIXTURE> 1 1.582144e-01
>  170552 <MEAN> 1
>  170553  -1.753086e-03
>  170554 <VARIANCE> 1
>  170555  8.353170e-04
>  170556 <GCONST> -5.249823e+00
>  170557 <MIXTURE> 2 8.417841e-01
>  ...
>
>  260387 ~h
> "CH_dz`-CH_U+sp/T:x_4_4_x_4/WS:1_6_6/CS:2_7_8/CW:2_1_2/PS:5_19_23/PW:5_1_5/PC:2_1_2"
>  260388 <BEGINHMM>
>  260389 <NUMSTATES> 7
>  260390 <STATE> 2
>  260391 <STREAM> 1
>  260392 ~p "mgc_s2_21"
>  260393 <STREAM> 2
>  260394 ~p "lf0_s2_523-2"
>  260395 <STREAM> 3
>  260396 ~p "lf0_s2_523-3"
>  260397 <STREAM> 4
>  260398 ~p "lf0_s2_523-4"
>  260399 <STATE> 3
>  260400 <STREAM> 1
>  ...
>
> There seem to be no error or in the stream[3]. What could affect the
> building process of regression tree?
>
> Regards,
> Jay
>
> 2012/2/24 Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
>>
>> Hi,
>>
>> The distributions are in .../cmp/re_clustered.mmf
>>
>> Regards,
>> Keiichiro Oura
>>
>>
>> 2012/2/24 li jay <lij.acd@xxxxxxxxx>:
>> > Hi,
>> >
>> > The following is part of the log file when I tried to build the
>> > regression
>> > tree:
>> >
>> > Splitting Node 32763, score 9.997541e+09
>> > (Stream=3, vSize=1)
>> > Splitting Node 32765, score 9.997541e+09
>> > (Stream=3, vSize=1)
>> > Splitting Node 32767, score 9.997541e+09
>> > (Stream=3, vSize=1)
>> > Splitting Node -32767, score 9.997541e+09
>> > (Stream=3, vSize=1)
>> > Splitting Node -32765, score 9.997541e+09
>> > (Stream=3, vSize=1)
>> > Splitting Node -32763, score 9.997541e+09
>> >
>> > The reason why the index went to negative value seemed to be an
>> > overflow occurred.
>> > Could you tell me in which file I can check the distribution of
>> > stream[3]?
>> > Thank you for you help.
>> >
>> > Regards,
>> > Jay
>> >
>> > 2012/2/23 Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
>> >>
>> >> Hi,
>> >>
>> >> This value is node index in "reg.tree".
>> >>
>> >>  printf("Splitting Node %d, score %e\n", r->nodeIndex, score);
>> >>
>> >> Index is always positive value.
>> >> I don't know why split is in a loop...
>> >> Could you check the distribution of stream[3]?
>> >>
>> >> Regards,
>> >> Keiichiro Oura
>> >>
>> >>
>> >> 2012/2/23 li jay <lij.acd@xxxxxxxxx>:
>> >> > Thank you for your reply.
>> >> >
>> >> > I've tried HTS2-2, replacing the HTS commands with the ones of
>> >> > HTS-2.2.
>> >> > The
>> >> > result was exactly the same as the one of HTS-2.1.1.  The score
>> >> > stayed
>> >> > the
>> >> > same, and the node split endlessly.
>> >> >
>> >> > What do you mean by 'Splitting Node' is negative value? You mean the
>> >> > score?
>> >> >
>> >> > Regards,
>> >> > Jay
>> >> >
>> >> > 2012/2/22 Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
>> >> >>
>> >> >> Hi,
>> >> >>
>> >> >> Could you try HTS-2.2?
>> >> >> I don't know why 'Splitting Node' is negative value.
>> >> >>
>> >> >> Regards,
>> >> >> Keiichiro Oura
>> >> >>
>> >> >> 2012/2/22 li jay <lij.acd@xxxxxxxxx>:
>> >> >> > Hi,
>> >> >> >
>> >> >> > I've been trying to build a regression tree for speaker
>> >> >> > adaptation. I
>> >> >> > am
>> >> >> > using HTS 2.1.1. I've trained a average voice model from 4000
>> >> >> > sentences
>> >> >> > (about 2.5 hrs) of 20 speakers. It was successful to generate
>> >> >> > voice
>> >> >> > using
>> >> >> > the average voice model. I wanted to apply speaker adaptation on
>> >> >> > this
>> >> >> > average voice model, so I tried to build a regression tree with
>> >> >> > the
>> >> >> > command
>> >> >> > below:
>> >> >> > /usr/local/HTS-2.1.1/bin/HHEd -A -B -C
>> >> >> > /home/jay/TTS/try/AST_female_20_speakers_2/configs/trn.cnf -D -T 1
>> >> >> > -p
>> >> >> > -i
>> >> >> > -H
>> >> >> > /home/ jay /TTS/try/AST_female_20_speakers_2/models/cmp/re_clust
>> >> >> > ered.mmf -M /home/
>> >> >> > jay /TTS/try/AST_female_20_speakers_2/models/cmp/regTrees
>> >> >> > /home/ jay /TTS/try/AST_female_20_speakers_2/edfiles/cmp/reg.hed
>> >> >> > /home/
>> >> >> > jay /TTS/try/AST_female_20_speaker
>> >> >> > s_2/data/lists/full.list
>> >> >> >
>> >> >> > The problem was that splitting of nodes did finish. It seemed to
>> >> >> > be
>> >> >> > in a
>> >> >> > loop, and the score stayed the same. So the HHEd command cannot
>> >> >> > stop.  The
>> >> >> > log file shows as below:
>> >> >> >
>> >> >> > HTK Configuration Parameters[10]
>> >> >> >   Module/Tool     Parameter                  Value
>> >> >> > #                 MINDUR                         5
>> >> >> > #                 MAXSTDDEVCOEF                 10
>> >> >> > #                 APPLYDURVARFLOOR              TRUE
>> >> >> > #                 DURVARFLOORPERCENTILE          1.000000
>> >> >> > #                 SHRINKOCCTHRESH  Vector 4 500.0 100.0 100.0
>> >> >> > 100.0
>> >> >> > #                 VFLOORSCALESTR  Vector 4 0.01 0.01 0.01 0.01
>> >> >> > #                 MINLEAFOCC                     0
>> >> >> > #                 NATURALWRITEORDER              TRUE
>> >> >> > #                 NATURALREADORDER              TRUE
>> >> >> > #                 APPLYVFLOOR                 TRUE
>> >> >> >
>> >> >> > // 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 10, score 8.998759e+10
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node 13, score 2.999760e+10
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node 4, score 1.000000e+10
>> >> >> > (MSD splitting)
>> >> >> > Splitting Node 15, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node 19, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node 21, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> > ...
>> >> >> > ...
>> >> >> > ...
>> >> >> > Splitting Node -16495, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node -16493, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> > Splitting Node -16491, score 9.997541e+09
>> >> >> > (Stream=3, vSize=1)
>> >> >> >
>> >> >> > Could you do me a favor to help the problem? My questions are:
>> >> >> > 1: What could be the reason or problem result in this endless
>> >> >> > splitting
>> >> >> > node
>> >> >> > situation.
>> >> >> > 2:Could it be the problem with the average modeling? Is there any
>> >> >> > option
>> >> >> > to
>> >> >> > enable average modeling? I trained the average model just as
>> >> >> > speaker
>> >> >> > dependent model with the same scripts, except the training data
>> >> >> > from
>> >> >> > different people.
>> >> >> >
>> >> >> > Thank you.
>> >> >> >
>> >> >> > Regards,
>> >> >> > Jay
>> >> >>
>> >> >
>> >>
>> >
>>
>



Follow-Ups
[hts-users:03186] Re: Problem of building regression tree (average voice model), li jay
References
[hts-users:03173] Problem of building regression tree (average voice model), li jay
[hts-users:03174] Re: Problem of building regression tree (average voice model), Keiichiro Oura
[hts-users:03176] Re: Problem of building regression tree (average voice model), li jay
[hts-users:03177] Re: Problem of building regression tree (average voice model), Keiichiro Oura
[hts-users:03179] Re: Problem of building regression tree (average voice model), li jay
[hts-users:03180] Re: Problem of building regression tree (average voice model), Keiichiro Oura
[hts-users:03181] Re: Problem of building regression tree (average voice model), li jay
[hts-users:03182] Re: Problem of building regression tree (average voice model), Keiichiro Oura
[hts-users:03183] Re: Problem of building regression tree (average voice model), li jay