[Dtb-talk] Converting tone indexed books to DAISY

[Aaron Cannon]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I agree that it would be a bad idea to play the cassette back at more than
2x speed.  What sampling frequency are you using for the initial recording?
I would recommend 44100, if that's not what you are already using.

There is however one very simple way to nearly cut in half your transfer
time.  If you, instead of using the LOC player, connected your IMic to a
standard stereo cassette deck, you could record two tracks at once.  If you
were recording side 1, you would have the audio for side at 2x speed on the
left track, and the audio for side 4 at 2x speed and reversed on the right
channel.  (On side 2, you will have side 2 on the left and side 3 on the
right reversed.)  Then, you would simply separate the two tracks, reverse
the right channel, and half the sampling speed, like you normally do.  Of
course, you'll need to use a player with good stereo separation, but that's
not all that uncommon.

As for finding the tones, I don't know what a good solution would be other
than the one you are using.  I don't think it would be that big a deal for
someone with the proper know-how to write a simple program to find all the
tones below 50HZ or so.

Good luck and keep us posted.

Sincerely
Aaron Cannon


- --
Skype: cannona
MSN/Windows Messenger: cannona at hotmail.com (don't send email to the hotmail
address.)
- ----- Original Message -----
From: "Greg Kearney" <gkearney at gmail.com>
To: "Technical Developments Discussion List"
<technical-developments at mail.daisy.org>; "Discussion of Digital Talking
Books" <dtb-talk at nfbnet.org>; "Kafka's Daytime" <contact at kafkasdaytime.com>;
"George Kerscher" <kerscher at montana.com>; "Mary Beth Janes"
<mjanes at apple.com>
Sent: Thursday, December 28, 2006 12:52 PM
Subject: [Dtb-talk] Converting tone indexed books to DAISY


>I have been experimenting with the process of converting a tone
> indexed tape book such as those from the National Library Service into
> DAISY setting the navigation at the indexed points. Here is what I am
> doing.
>
> Hardware: For the playback I am using a standard NLS C-1 tape player.
> I am recoding into an Apple Macintosh computer using an iMic USB audio
> interface. I patch between the AUX out port on the player into the
> line in jack of the iMic. I have a Maxstore Firewire drive which is
> used as the scratch disk for the recording software and as the
> location for the audio files.
>
> Software: For software I am using Sound Studio 2.5. This software will
> record to disk in the background. It will also let me stretch out the
> sound, apply filter to reduce the tape hiss and most importantly mark
> and then split the recording at the marked points.
>
> Process:
>
> 1. Record each side of the book into a AIFF file at 1 7/8 speed. This
> is twice the normal payback speed and each track records at about 1
> hour. The recoding is done as Mono as there is no stereo information
> coming from the playback device.
>
> 2. Once the recording is finished I apply filter to reduce hiss and
> then "stretch" the recoding by changing the sample rate to twice the
> time as the original recording as it was recorded at 2X speed. I now
> have a normal speed audio file.
>
> 3. Tone indexing appears as a low frequency signal just before the
> start of the content on the visual display of the recording. I simply
> scroll through the recording looking for this distinctive visual
> signature. (I'm a dyslexic so looking for the visual signature is
> simple for me.)
>
> 4. At the point just beyond the index tone I place a marker using the
> marker function of the Sound Studio software. I also name the marker a
> meaningful name such as contents, preface, chapter 1 and so on.
>
> 5. Once the marker are set and named I then use Sound Studio to split
> the recording apart at the marker giving me a directory filled with
> recording. I can then use these in my DTBmaker program to generate a
> DAISY DTB  using the set of recording.
>
> I could improve this process by having a player that could play at 3X
> or 4X speed which would quicken the production. However I know of no
> such device. It would be interesting to see if there were a way for
> the computer to detect the index tones and mark the file. Then you
> would only need to check that the markers were in the right places in
> the file. This would permit the blind to do the conversion process as
> well. Further it would be useful if the National Library Service had
> tone indexed their book at the page level rather than just at the
> chapters. The DAISY book I was able to produce with this method is
> navigable to the chapter only.
>
> Greg Kearney
> Wyoming Medical Center
> 307-577-2473
> _______________________________________________
> Dtb-talk mailing list
> Dtb-talk at nfbnet.org
> http://www.nfbnet.org/mailman/listinfo/dtb-talk
>
>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (MingW32) - GPGrelay v0.959
Comment: Key available from all major key servers.

iD8DBQFFlCLDI7J99hVZuJcRAloiAJwJxVtNcYD61tR75Sw5wzL9shzTfgCg0YnM
y70UOG+DjTX+OwYoqD6Omms=
=4tgI
-----END PGP SIGNATURE-----