Subject: Re: [linux-audio-user] Piping multiple text documents into Audio
From: Janina Sajka (janina_AT_rednote.net)
Date: Mon Dec 06 2004 - 16:09:11 EET
mark_AT_kfm.co.za writes:
> Hello Everybody
>
> I have a lot of text documents that I wish to output into speech
> audio. Firstly I wanted to know if there is any software that would
> allow one to do this.
You probably have, or could easily get, Festival on your Linux system.
It's engineered specifically for such tasks. The man page should get you
started.
If you want another, look at:
Both Festival and FreeTTS support the Mbrola plugins, so you can get
utterance in various languages.
There are other, higher quality but proprietary, Text To Speech (TTS)
engines that might meet your needs better, especially if you or your
users are new to understanding the mechanical utterance characteristics
of synthetic speech engines. Take a look at:
http://www.cepstral.com
http://www.fonix.com
Both of these companies offer Linux TTS engines, as does AT&T for whom I
have only their demo URL:
http://www.research.att.com/projects/tts/demo.html
Also, look at:
http://trace.wisc.edu/docs/speechsamples/index.htm
> This could be likened to an educational tool
> of sorts that reads a file.
>
Hardly. Reading from beginning to end is just the first step in turning
TTS into a tool. There's much much more to it. If you're interested in
the state of the art, however, you should look at DAISY:
> Secondly I was hoping to change the output of audio by the
> actual content of the text/html page. The actual reading of text
> would be at quite a speed (not necessarily comprehendible) but
> in essence the content itself would create the music - according
> to an adjustable tempo/spec? Maybe each letter could represent
> a music note?
You can certainly create any kind of mapping you might want. Perhaps the
SSML spec might help you?:
http://www.w3.org/TR/2004/REC-speech-synthesis-20040907
>
> I am not sure if this makes sense but if possible I would like to
> hear if there are any comments?
>
> MAny thanks for your time and have a good week!
>
> CheersRegards
> Mark
>
> Mark McBride
> 0844146809
>
>
>
> ____________________________________________________________________________
> The information in this e-mail is confidential and is intended solely for
> the addressee. Access to this e-mail by anyone else is unauthorised. If
> you are not the intended recipient, any disclosure, copying, distribution
> or any action taken or omitted in reliance on this, is prohibited and may
> be unlawful. Whilst all reasonable steps are taken to ensure the accuracy
> and integrity of information and data, and to preserve the confidentiality
> thereof, no liability or responsibility whatsoever is accepted if
> information or data is corrupted or does not reach its intended destination.
> KFM Radio (Pty) Ltd. will not accept responsibility for unauthorised use,
> be it public or private, to express opinion, promote or demote individuals
> or groups.
-- Janina Sajka, Chair Accessibility Workgroup Free Standards Group (FSG)janina_AT_freestandards.org Phone: +1 202.494.7040
This archive was generated by hypermail 2b28 : Mon Dec 06 2004 - 16:18:36 EET