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

[hts-users:00777] Re: Questions on training and flat pitch pattern


I have another tricks to reduce memory consumption in training process.
But, the tricks would confuse other HTS-users so much.

I would like to discuss this point with Dr.Zen and other members
the week after next. Hopefully, I would like to introduce a better solution
to cope with this problem.

By the way, do you have 64-bit linux machines?
Using the 64-bit linux machines is a quickest way to solve this problem.

Junichi


On 2007/08/09, at 14:10, Lee Sillon wrote:

Could you tell me how to modify it? thanks in advance!

2007/8/7, Junichi Yamagishi <jyamagis@xxxxxxxxxxxx>: Hi,

On 2007/08/07, at 16:45, Heiga ZEN (Byung Ha CHUN) wrote:

> Lee Sillon wrote (2007/08/08 0:10):
>
>> 1. If the training corpus is larger (for example,10,000
>> sentences), the training process will crash because of memory
>> consuming(about 2GB). Could it be solved?
>
> I guess tree-based clustering (HHEd) consumes huge memory for
> larger training data. You can use low-memory implementation of tree-
> based clustering by specifying -r option.

I've trained HMMs for HMM-based speech synthesis from 23,000 sentences.

The HTS toolkit worked well.
(although I needed to slightly modify training scripts
for running the HTS commands.)

Regards,
Junichi








--
Have a nice day!
Sillon Lee

welcome to: http://spaces.msn.com/zouzoumao/


References
[hts-users:00764] Questions on training and flat pitch pattern, Lee Sillon
[hts-users:00765] Re: Questions on training and flat pitch pattern, Heiga ZEN (Byung Ha CHUN)
[hts-users:00766] Re: Questions on training and flat pitch pattern, Junichi Yamagishi