[Bookreader] [IAEP] Text to Speech readers for XO
mccabe at archive.org
Tue Nov 3 17:50:45 EST 2009
Hi all -
I'm working on creating DAISY (and epub) books for the Archive, and I'm
very happy to answer any questions. DAISY and epub books have many
Sayamindu Dasgupta wrote:
> The Internet Archive has started to distribute books as DAISY
> (http://en.wikipedia.org/wiki/DAISY_Digital_Talking_Book), something
> we should definitely take a look at. We might also consider leveraging
> the GNOME accessibility framework to provide book-reading features for
> Epubs and PDFs in Read - it may be tricky, but the end results would
> be worth it.
> On Fri, Oct 30, 2009 at 5:34 AM, Samuel Klein <meta.sj at gmail.com> wrote:
>> Bumping up this recent thread on the bookreader list about text-to-speech.
>> Mike and Gregor, in case you haven't seen what's currently possible:
>> I believe James S's Read Etexts uses speech-dispatcher to read selected
>> text. Aleksey and others may have done further work with espeak... I've
>> included some old threads from the Sugar list this past spring below.
>> On Thu, Oct 29, Mike McCabe <mccabe at archive.org> wrote:
>> I also think this is a great idea. I've worked with several
>> text-to-speech readers recently, as part of my effort to make the
>> Internet Archive books available to print disabled people.
>> They're very useful, and I think that this mode of reading could be of
>> use to a very broad range of users. I suspect we'll see more of it soon.
>> I'm also curious to hear about specific experiences with
>> linux-compatible free TTS, as we may be producing audio books with this
>> to work with the new Library of Congress audio players.
>> Best regards -
>> ==  old note from James Simmons ==
>> ( in repsponse to this speech-synthesis summer of code proposal:
>> http://wiki.sugarlabs.org/go/speech-synthesis )
>> Since you have been working with Aleksey Lim you probably know about
>> text to speech with highlighting in Read Etexts. I wrote the original
>> TTS code that used speech-dispatcher with some assistance from Hemant
>> Goyal and the folks on the speech-dispatcher project. Aleksey
>> refactored my code so it could work with either speech-dispatcher or his
>> own gstreamer espeak plugin. Not only does his plugin need no
>> configuration to work, it also does a LOT better in producing timely
>> callbacks as it reads each word.
>> As you point out in your proposal, highlighting the word as it is spoken
>> is a big part of the benefit of what you're proposing. If all you
>> wanted to do was capture some highlighted text in the clipboard and have
>> it spoken in a voice you can configure in a control panel, that would be
>> easy, even trivial. It's the highlighting that's difficult. When I
>> added speech to Read Etexts I deliberately tried for the simplest
>> approach that would get the job done. It reads only the current page.
>> It always starts either at the first word on the page, or if speech has
>> been paused, it resumes with the last word spoken. You can't choose the
>> word to start on. The Activity itself receives the callbacks as each
>> word is spoken and takes care of doing the highlight and scrolling the
>> textarea so the highlighted word stays on the screen.
>> If I had to write a facility that did what Read Etexts does outside of
>> the Activity I wouldn't know how to do it. It seems to me that
>> highlighting is best done by the Activity itself. I can't deny that it
>> would be useful to have all this work done as you have described without
>> the Activity knowing anything about it, but it doesn't seem feasible.
>> You'd have to have something that could work with gtk textareas, the
>> evince component Read uses, Abiword, and everything else that came along.
>> Another thing you'd have to deal with is PDFs composed of scanned in
>> book pages. There are a lot of these around (the Internet Archive is
>> full of them) and somehow the kid trying to select words on a scanned in
>> page would have to be clued in that these words are not selectable.
>> I suppose you could make an Activity that grabbed whatever text was in
>> the clipboard, displayed it in a textarea, and highlighted the words in
>> that textarea as it spoke them. I'm pretty sure that wasn't what you
>> had in mind.
>> Splitting sentences into separate words will be a challenge. I just use
>> spaces as delimiters and filter out characters like asterisks, vertical
>> bars, etc. That works OK for English but not for other languages. If I
>> wanted Read Etexts to do highlighting on the Bhagavad-Gita in the original
>> Sanskrit it wouldn't work. Even in English I get tripped up by double
>> hyphens (--). It would be nice if Gutenberg etexts put spaces around double
>> hyphens but they don't.
>> It looks like you've picked a challenging project, and I would love to be
>> proven wrong about everything I've mentioned here. Good luck with this,
>> James Simmons
>> == 2: SynPhony and reading assistance ==
>> On Tue, Feb 17, 2009 at 12:48 PM, Carol Farlow Lerche <cafl at msbit.com>
>>> I'd like to call your attention again to SynPhony. We are close to a base
>>> release (probably this week) of a 44,000 word English word database that has
>>> a very rich array of information helpful to the teaching of English,
>>> especially reading. A 10,000 word Spanish lexicon and 50000 word German one
>>> will follow. Norbert Rennert who compiled these, would like very much to
>>> work with other language experts to extend this effort to other languages.
>>> Some highlights of the English lexicon: screened from the CMU Sphynx corpus
>>> for accessibility to children, each word entry has frequency data from
>>> analysis with respect to a large corpus of text merged in, phoneme breakdown
>>> (used by reading curricula to decide the order in which words should be
>>> introduced or deemed decodable), etymology, semantic domain
>>> (categorization), IPA coding, syllabification and stress marking.
>>> The second release will merge in many images, though we don't expect to
>>> have a complete image-to-word mapping without a volunteer effort. We plan
>>> to create an API and a way to define a curriculum sequence for word groups
>>> once the basic database is released, to allow integration of the word bank
>>> across all the activities that are literacy related, as well as create
>>> more. We also hope to use the word bank to score texts for reading level
>>> and assist in creation of simplified version of extant texts suitable for
>>> use by emergent readers. Please read our design documents at the above
>>> On Tue, Feb 17, 2009 at 2:02 AM, Tomeu Vizoso <tomeu at sugarlabs.org> wrote:
>>>> Aleksey has started a very interesting new path:
>>> Gregor Kervina wrote:
>>>> Hi Sayamindu,
>>>> thanks for quick reply!
>>>> There is a lot of text to speech software out there - I use
>>>> http://www.bytecool.com/coolspch.htm that you can try trial and download
>>>> additional voices, just to get a feeling, but it is not free and not for
>>>> linux. Many other programs are more complex and complicated and some of
>>>> them use very complex voice engines that in my opinion doesn't sound
>>>> very good. (I use Mary voice with cool speech)
>>>> OK I spent some time to find all TTS software that is free for linux and
>>>> here are some links:
>>>> http://larswiki.atrc.utoronto.ca/wiki/Software - see the links under
>>>> Speech section
>>>> http://www.cstr.ed.ac.uk/projects/festival/onlinedemo.html - listen to
>>>> some demo voices
>>>> http://sourceforge.net/projects/dhvani/ - this one not english
>>>> http://www.speech.cs.cmu.edu/hephaestus.html - click the links in Speech
>>>> Synthesis section
>>>> http://www.linux.com/archive/feature/122197 - two readers - plug-ins for
>>>> I can not test them because I'm not a linux user. Maybe you can modify
>>>> some of these software (probably Festival) for more user friendly
>>>> reading and maybe program a specific button on XO keyboard that will
>>>> automatically read the selected text no matter what program is used for
>>>> opening the text.
>>>> Judging from google search result for DTBooks, this technology is not
>>>> spread at all. The other problem is that it uses somtimes recorded audio
>>>> and the size of that is too large for XO... I think the most important
>>>> is that TTS works with reader that will open 1.6M e-books from internet
>>>> you in this team?).
>>>> Also one important thing is to add cheap headphones with laptop so
>>>> children could listen to reading without desturbing others and in the
>>>> noisy environments ... another advantage of audio reading is much longer
>>>> battery life because you can turn off LCD monitor and audio alone does
>>>> not consume much energy.
>>>> Let me know what you think.
>>>> All the best,
>>>> On Mon, Oct 26, 2009 at 4:08 PM, Sayamindu Dasgupta <sayamindu at gmail.com
>>>> <mailto:sayamindu at gmail.com>> wrote:
>>>> Hi Gregor,
>>>> Thanks a lot for jumping in :-)
>>>> On Mon, Oct 26, 2009 at 2:38 AM, Gregor Kervina
>>>> <gregor.kervina at gmail.com <mailto:gregor.kervina at gmail.com>> wrote:
>>>> > Dear Sayamindu Dasgupta, SJ Klein and other members of this list,
>>>> > I'm a student of electrical engineering from Europe and would
>>>> like to share
>>>> > with you my very positive experience with text to speech
>>>> technology that can
>>>> > in my opinion significantly increase the educational potential of
>>>> XO if used
>>>> > in the right way.
>>>> > For the past 12 years (since I was 15 years old) I'm daily
>>>> learning from
>>>> > e-books and internet using text to speech software. I know this
>>>> software is
>>>> > unpopular in developed world, many people don't even know that it
>>>> exists. On
>>>> > the other hand many people (including me) don't like reading long
>>>> texts on
>>>> > the LCD screens - that's why e-books are also not very popular.
>>>> > But unlike my friends I read 50+ e-books every ear and also daily
>>>> news on
>>>> > the internet - I just select the text, copy it, and CoolSpeech
>>>> > (using Mary voice) reads me all the text with speeds 300 to 500
>>>> words per
>>>> > minute. In this way I can browse other sites or look at photos or
>>>> just lay
>>>> > down and listen while my laptop is reading to me.
>>>> > Other people don't understand what I'm reading because it is too
>>>> fast for
>>>> > them but it can be learned quickly with slower speeds at
>>>> > I think XO laptops should definitely have such software
>>>> pre-installed and a
>>>> > video introduction how to use it and what reading speeds can they
>>>> > after some time of practicing.
>>>> > It is also ideal for children with poor eye sight.
>>>> This sounds awesome. Could you let us know if the text to speech
>>>> software you have in mind is free/opensource and if it works on
>>>> I am also looking at DTBooks specifications for digital talking
>>>> - do you know how useful/widespread this technology is ?
>>>> Sayamindu Dasgupta
>>>> Bookreader mailing list
>>>> Bookreader at lists.laptop.org
>>> Bookreader mailing list
>>> Bookreader at lists.laptop.org
>> IAEP -- It's An Education Project (not a laptop project!)
>> IAEP at lists.sugarlabs.org
More information about the Bookreader