[vlc-devel] Need help developing text-to-speech-module in VLC - offering £5.000

[Rémi Denis-Courmont]

   Hej Sandra,

On Thu, 7 Apr 2011 10:43:31 +0200, Sandra Derbring
<sandra.derbring at gmail.com> wrote:
(...)
> We work from the premise that currently, these people can't take part of
> foreign movies where you have to read the subtitles to follow what's
> happening in the movie. We want to add a module to VLC that can call on
a
> speech synthesis on the user's system and sync the reading with the
> subtitles. Basically, instead of hearing the original voices, you hear
the
> Swedish (or preferred language) speech syntheis voice. We'd like the
> subtitles to be displayed as before. We'd also want this function to
> operate cross-platform like the rest of VLC.

You need to acquire the subtitles first. VLC supports roughly four classes
of subtitles:

 - plain text subtitles (e.g. digital TV subs and *.srt files),
 - rich subtitles (e.g. SSA),
 - bitmaps (e.g. DVD),
 - "burnt" subtitles.

Voice can be synthesized from plain text directly, and from rich text at
the cost of loosing the "enrichment". But to read bitmaps, you would need
optical characters recognition. If that is required (especially for DVD
subtitles playback), there may be a very hard problem. As for "burnt"
subtitles, ouch, I guess OCR would be very hard and unreliable.

Then you need a speech synthesis engine. There are quite a few open-source
ones. But I don't know which ones, if any, supports Swedish phonetics and
have a GPL-compatible copyright license. Did you already sort out that part
of the equation?

And last, you need to filter out the original voice from the original
audio track. Or do you not mind loosing the original audio sound effects? I
don't suppose you will always have a clear/speech-less audio channel
available in the original media.
 
> We're just in the middle of applying for funding for our project and
> need to know if this could be a way for us to go - that is, if there
> is any interest from anyone to do this. We are also interested in
> knowing the amount of time that would be required to develop this
> feature - does anyone have an approximation?

It depends a lot on the requirements, and what existing components could
be sourced or would be provided by you.

Best regards,

-- 
Rémi Denis-Courmont
http://www.remlab.net/