DemiPuppet Newbie United States Joined 6587 days ago 19 posts - 55 votes
| Message 1 of 5 16 February 2015 at 5:10pm | IP Logged |
Once again I did some more digitizing of missing FSI language audio from official US government tapes using Audacity. The More Basic course was missing units 29 - 48. I've digitized those as well as units 1 - 28. I also ran an OCR over the text and appended the original cover. Those, as well as pictures of the US government tapes used for digitizing, are in the ~1.0GB file below:
kanbox.com/f/v70Mg
As you know, per US copyright law, works created by US federal employees as part of their normal job are not eligible for US federal copyright.
As before, my only changes were to do a very slight noise reduction (Audacity Noise Removal Settings: Reduction Amount 3dB, Sensitivity 0dB, Freq Smoothing 250Hz, Attack/Decay 100ms) and normalizing of the volume (Audacity setting -1dB, each channel done independently). I encoded the16 bit audio recorded via Audacity with the Lame MP3 encoder set to variable bit rate V4 with joint stereo (approx 140kbps).
An MP3 encoder setting of V4 has an upper frequency limit of 17.5KHz. This really makes no sense at first glance since standard non-Dolby cassette audio only goes up to about 14KHz and the human voice only goes up to about 12KHz. Also, 16 bit recording makes no sense since standard cassette tape only has about a 12bit resolution. My goal is not to produce the smallest best sounding versions of the files, but rather to reproduce the audio as closely as possible (with only very minor changes) to the original version on the tape. It is up to the users to optimize to their preferences. Therefore I haven't done any extra equalization or bandpass manipulation of the files.
I would suggest re-encoding at a lower encoder setting such as V7 which has an upper limit of 14.75KHz. That would remove some of the high end noise. This can be done with Audacity using their "chains" batch processing. This reduces the size, retains the M3 tags, and probably produces a better sounding file. The batch feature also allows applying an equalization curve against the audio, though one needs to be careful not to over amplify and possibly cause clipping. Memory is cheap; always keep a copy of the original files as a backup.
4 persons have voted this message useful
|
Ericounet Senior Member France yojik.euRegistered users can see my Skype Name Joined 5435 days ago 157 posts - 414 votes Studies: English, German, Russian
| Message 2 of 5 16 February 2015 at 5:27pm | IP Logged |
hi
thanks for your work .. downloading the material at the moment.
I'll put on the website as soon as possible.
Eric
1 person has voted this message useful
|
tarvos Super Polyglot Winner TAC 2012 Senior Member China likeapolyglot.wordpr Joined 4707 days ago 5310 posts - 9399 votes Speaks: Dutch*, English, Swedish, French, Russian, German, Italian, Norwegian, Mandarin, Romanian, Afrikaans Studies: Greek, Modern Hebrew, Spanish, Portuguese, Czech, Korean, Esperanto, Finnish
| Message 3 of 5 16 February 2015 at 6:44pm | IP Logged |
What language is this?
1 person has voted this message useful
|
DemiPuppet Newbie United States Joined 6587 days ago 19 posts - 55 votes
| Message 4 of 5 16 February 2015 at 7:08pm | IP Logged |
This is the "Mossi" west African language. It is also called "More".
http://en.wikipedia.org/wiki/Mossi_language
1 person has voted this message useful
|
Ericounet Senior Member France yojik.euRegistered users can see my Skype Name Joined 5435 days ago 157 posts - 414 votes Studies: English, German, Russian
| Message 5 of 5 22 February 2015 at 11:10am | IP Logged |
hi,
new DemiPuppet material integrated on the website.
Thanks a lot, DemiPuppet
Eric
1 person has voted this message useful
|