[hts-users:03799] Re: How to convert hts_voice 1.03 to new format (1.05)?

[李贤 <shysian@xxxxxxxxx>]

and modify the make_htsvoice() function

1. HHEd shoud be HTS 2.3 version
2. use make_lpf() function in HTS-2.3 alphs demo 

Regards,

Li Xian

2013/6/20 Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>

Hi,

Please, check make_htsvoice() function in scripts/Training.pl (HTS-2.3
alpha demo script).

Regards,
Keiichiro Oura



2013/6/20 fishman8.tw <fishman8.tw@xxxxxxxxxxxx>:

> Hello,
>
> I had compile the hts_voice_nitech_jp_atr503_m001 retrieved from HTS
> website.
> Finally it produced :
>
> dur.pdf
> gv-lf0.pdf
> gv-mgc.pdf
> gv-switch.inf
> lf0.pdf
> lf0.win1
> lf0.win2
> lf0.win3
> lpf.pdf
> lpf.win1
> mgc.pdf
> mgc.win1
> mgc.win2
> mgc.win3
> tree-dur.inf
> tree-gv-lf0.inf
> tree-gv-mgc.inf
> tree-lf0.inf
> tree-lpf.inf
> tree-mgc.inf
>
> these files. But the new voice format will combined them into one file
> called .htsvoice .
> We can download the compiled 1.05 file and get
> nitech_jp_atr503_m001.htsvoice .
> Since hts_engine 1.07 / open jtalk 1.06 using the new format , how can I
> convert my voice files into new one?
>
>
> Thank you.