[Project-ideas] GSOC 2013 Introduction

[Bhavani Shankar R]

On Fri, Apr 12, 2013 at 8:20 AM, Sankarshan Mukhopadhyay
<sankarshan.mukhopadhyay at gmail.com> wrote:

> I am not familiar with the modifications required in Dhvani to make it
> work as a TTS, I am aware that Dhvani may be a good choice. Perhaps
> Bhavani could provide an opinion.

Hi Sankarshan,

Apologies for the delayed response.

Dhvani is a good choice and I second it but I am not sure about
bengali but in kannada and telugu I think I faced some problems in
syllabification of simple words and I am not sure whether its a
problem with my system or the software. For ex:

ಹಲೋ (English: Hello) when I try to decode it comes as ಹ,ಲೊ,ೕ with
last charecter being scrambled

and in telugu too it works fine except facing similar kind of issue for example

మీరు ఎక్కడ ఉన్నారు? (English: where are you?) when I try to decode in
my system, it comes as

మ,ీ,ర,ు, ,ఎ,క,్,క,డ, ,ఉ,న,్,న,ా,ర,ు,?

with scrambled charecters again..

Not sure its a problem with the software or the system as I said
earlier and these are few of my experiences.

I agree that Dhvani is a good choice but needs some improvement I
guess (It might be worth to note that the last version of Dhvani  got
released in Nov 2008 IIRC and improving the same can be a great
project I think :-) )

Regards,


-- 
Bhavani Shankar
Ubuntu Developer       |  www.ubuntu.com
https://launchpad.net/~bhavi