My 1-Minute Test (listeners wanted) [OK]

All languages: post your test recording here. Help check audio files, provide editing services, and advertise for proof-listeners.
Post Reply
Lushnam
Posts: 115
Joined: December 13th, 2008, 12:32 am
Location: Idaho
Contact:

Post by Lushnam » January 9th, 2019, 8:26 pm

Here's a link to the 1-minute test recording that I did: https://librivox.org/uploads/tests/test_lushnam.mp3

I used the black, 16GB digital voice recorder that I found here (using the built-in microphone which is louder than the external one it comes with): https://www.amazon.com/gp/product/B07KBWN8L1/

I recorded in the highest quality (1536kbps) WAV format (I didn't know they had bitrates) and then imported it into Audacity 2.2.1 (on Xubuntu 18.04, 64-bit, Linux); then I removed the start/stop button-press sounds at the beginning and end (the built-in microphone records the stop/pause button presses very loudly, although it doesn't usually record a button sound at the beginning), and exported it to the desired constant 128kbps mono MP3 format.

I've tried recording with the device a foot and a half away and with it about a foot away, and the volume is about the same either way. My recording here is the closer one. MP3Gain says it has a dB of 88.4 (MP3Gain works with the Windows emulator on Linux).

My recording passed the checker (Linux version): https://cgjennings.ca/checker.html

I ordered a webcam from BestBuy (one that is supposed to have a good microphone), which I can hopefully also use (for recording audio) in addition to this device, but I'll have to try that out later (when it arrives).

My digital voice recorder's sample rate is 48,000hz instead of 44,100hz (for both my MP3 and the original WAV file). I've figured out how to change that in Audacity without lowering the pitch of my voice (there's a bar at the bottom of the screen where I can change the project rate, in hz; then, when I export it from WAV to MP3, it has the right sample rate). So, the MP3 in the 1-Minute Test link above is now at 44,100hz instead of the 48,000hz. I understand that changing it to 44,100hz should lessen the quality to a probably unnoticeable degree, but that it should make the MP3 workable on more devices.

However, I'm still confused about the sample size (the wiki calls for 16-bit). What is that called in Audacity? Is it the format? If so, Audacity seems to default to 32-bit float for everything (even if the original file was 16-bit), but changing it in the settings (Edit->Preferences->Quality->Default Sample Format) seems to change that default. My original WAV file is 16-bit PCM.

My digital voice recorder itself has no settings for changing sample rate/size, as far as I've discovered. So, that's why I changed the sample rate in Audacity.

Anyway, thanks for the help you provide, and please let me know what else I should do before I start doing LibriVox recordings.

Kitty
LibriVox Admin Team
Posts: 17285
Joined: March 28th, 2014, 5:57 am

Post by Kitty » January 10th, 2019, 3:26 am

hi Mark :) I saw your reply on the other test. Thanks for helping out here. I see you are a returning volunteer, so you know your way around the forum already ;)

I also checked your test just now and your reading is absolutely fine too. The tech specs are correct. I don't really know about the sample size, I need to check what I have in my Audacity program. But the Checker program gives it a pass, so all is well I suppose :lol:

The sample rate indeed needs to be 44,100 Hz and it can happen that the voice gets distorted if you change it from 48,000. I had that problem once with a reader. So best to record at 44,100 from the start if possible.

The only thing that was a bit disturbing was the background noise. I saw in the other test thread that you know how to noise-clean in Audacity, so I will not hold you back for that. But it would definitely be a good idea to clean your official sections up with one round of noise-cleaning before submitting. It makes the listening experience a bit more agreeable and it would be a shame to mar it just because of this tiny thing. The reading otherwise was well done. :thumbs:

You can reupload a cleaned version if you wish me to have another look at it :) Then post in this same thread, so I get notified.

But: here's your OK, you're free tuck into a project on offer.

have fun !

Sonia

Lushnam
Posts: 115
Joined: December 13th, 2008, 12:32 am
Location: Idaho
Contact:

Post by Lushnam » January 10th, 2019, 4:03 am

Thanks Sonia (and you're welcome)! :)

Yeah, I was wondering if I should reduce the noise or not.

I reduced the noise and uploaded the new file (same name). These are the settings I used:

Noise reduction (dB): 8 (the default of 12 made it sound synthetic a bit)
Sensitivity: 6.00 (this was the default for noise reduction)
Frequency smoothing (bands): 3 (this was the default for noise reduction)

Kitty
LibriVox Admin Team
Posts: 17285
Joined: March 28th, 2014, 5:57 am

Post by Kitty » January 10th, 2019, 4:06 am

Lushnam wrote:
January 10th, 2019, 4:03 am
Yeah, I was wondering if I should reduce the noise or not.
I reduced the noise and uploaded the new file (same name). These are the settings I used:

Noise reduction (dB): 8 (the default of 12 made it sound synthetic a bit)
Sensitivity: 6.00 (this was the default for noise reduction)
Frequency smoothing (bands): 3 (this was the default for noise reduction)
to my ear the sound is fine now :) I definitely think you can use those settings. Every microphone or computer needs different settings, so you'll hear by trial and error what will work for you.

For me, this is a really good recording. You're totally ready to start :)

Have fun

Sonia

Lushnam
Posts: 115
Joined: December 13th, 2008, 12:32 am
Location: Idaho
Contact:

Post by Lushnam » January 10th, 2019, 4:09 am

Awesome! Thanks again! Now I can get started on helping with The Chinese Kitten (section 10).

Post Reply