From: ds2list <ds2list_at_yahoo.com>
Date: Wed, 07 May 2003 02:13:35 -0700

i don't know much about text-to-speech libraries, but that might
be a bit of a heavy solution for Rockbox. Already the binary is
very close to the 200K limit (Talkbox is 198.something K). But
then again, I've never used a text-to-speech system. Someone
on the list a few days ago mentioned something about moving to
a plugin architecture. Maybe that will allow text-to-speech
in the future.

-danan

On Wed, 07 May 2003 09:05:55 +0200, Holger Lembke wrote:

>> You can either create these folder name mp3s yourself and put them
>> in the appropriate folder, or you can press F2-> Record Thumbnail
>> while in the appropriate folder, to record it directly on the
>> Recorder via the Mic/Line-In jack.
>
> Perhaps you don't know this link, so...
>
> ---------------------------
> http://tcts.fpms.ac.be/synthesis
>
> MBROLA is a speech synthesizer based on the concatenation of
> diphones. It takes a list of phonemes as input, together with prosodic
> information (duration of phonemes and a piecewise linear description
> of pitch), and produces speech samples on 16 bits (linear), at the
> sampling frequency of the diphone database.
> ---------------------------
>
>
>
> --
> Holger Lembke Keep your Mailbox SPAM free:
> +49-531-334676 Disruptor OL http://disruptor.de/
Received on 2003-05-07