Here's a 1-minute test! [OK]

All languages: post your test recording here. Help check audio files, provide editing services, and advertise for proof-listeners.
Post Reply
nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » September 1st, 2019, 3:15 pm

Hello! here is a test for feedback on my recording situation. :D

recording software - Audacity
operating system - Windows
mic type - Analog

https://librivox.org/uploads/tests/librivox1mintestnggillespie.mp3

Kitty
LibriVox Admin Team
Posts: 22008
Joined: March 28th, 2014, 5:57 am

Post by Kitty » September 2nd, 2019, 10:50 am

Hi Nathaniel and welcome to Librivox

thank you for starting with a test, always good to have the technical things ironed out from the start ;) there are a few technical issues, but first and foremost, your reading proper is absolutely wonderful to listen to. Nice enunciation, good pacing, easy to listen to. :thumbs: So far so good.

On the technical side, we need a few more tweakings though to make this standard LV. You have Audacity, so that's good, I can give you easy step-by-steps. No need to re-record anything, just correct your current test and all is well.

1) your volume is too low. We accept between 86 and 92 dB, yours is only 84.4. To increase volume, do the following:

> highlight your recording
> go to Effect
> go to Amplify
> input 5 (to get to a good average of 89) in the first upper box (called "amplification dB")
> press ok

2) your DC bias is too high, this should only be around -1, 0 or +1, yours is -125 :shock: you can normalize the DC offset easily:

> highlight your recording
> go to Effect
> go to Normalize
> tick Remove DC offset box and press OK

3) your bitrate is variable instead of 128kbps constant. This gets set while exporting to mp3:

> go to File / Export Audio
> open Options in the lower right corner
> select the following settings: Bit Rate Mode: Constant / Quality: 128 kbps
> press Ok

and that's it. Could you change these three settings please and reupload your corrected test-file in this same thread so I get notified and can have another look ?

Thanks :)

Sonia

nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » September 3rd, 2019, 12:16 pm

Awesome! I'll fix those issues and get back to you. Thank you!

nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » March 24th, 2020, 12:03 pm


Kitty
LibriVox Admin Team
Posts: 22008
Joined: March 28th, 2014, 5:57 am

Post by Kitty » March 24th, 2020, 12:41 pm

:shock: it's been a while since I heard back from you, Nathaniel

thanks for the second test. Now the bitrate at least is correctly set :thumbs: so you need not change anything there anymore

The volume strangely enough is now over our accepted limit (over 92 dB). Did you re-record or did you amplify like I said ? With only +5 it should not have exceeded that much.

Could you try decreasing the volume this time ? No re-recording, use this same test please:

> highlight your recording
> go to Effect
> go to Amplify
> input -2 (don't forget the minus sign !) in the first upper box (called "amplification dB")
> press ok

Also the DC bias is still higher than last time, (-308). Did you do the DC removal, like I told you before ? :hmm:

> highlight your recording
> go to Effect
> go to Normalize
> tick Remove DC offset box and press OK

Could you please try once more correcting these two points, then all is well. If you have any problems with the program, please tell me exactly what you did and maybe we'll find a solution together

Don't give up, we'll get to the bottom of it :)

Sonia

nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » March 24th, 2020, 1:23 pm

yeah, it has been a while.

I didn't rerecord anything, I tried my best to make the changes you were referring to

i think i did forget the minus sign

also, remove dc offset was already checked so I unchecked the box thinking that was maybe what you meant?

Kitty
LibriVox Admin Team
Posts: 22008
Joined: March 28th, 2014, 5:57 am

Post by Kitty » March 25th, 2020, 7:57 am

nathanielgillespie wrote:
March 24th, 2020, 1:23 pm
i think i did forget the minus sign
no back in my first post you had to amplify, so it was correct with the plus sign. But somehow it still was too loud in the end which it shouldn't have been. Did you write the number in the correct box ? The upper part, not the lower part. Now this time, since it it too loud, you need to add the minus sign.
also, remove dc offset was already checked so I unchecked the box thinking that was maybe what you meant?
ah yes, sorry, the box has to be checked. If it was already checked, then leave it and simply press OK. I sometimes uncheck and re-check the box, just to make sure the programme gets the hint :lol:

Sonia

nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » March 25th, 2020, 1:44 pm

Oki-doki

Hopefully this one is better, it doesn't sound much quieter but the soundwave shrank so i guess that's a good sign

https://librivox.org/uploads/tests/librivox1mintestnggillespie3.mp3

Kitty
LibriVox Admin Team
Posts: 22008
Joined: March 28th, 2014, 5:57 am

Post by Kitty » March 26th, 2020, 3:25 am

excellent, yes this is right inside our volume limits. :thumbs: The DC bias is still very high, but it doesn't distort the sound, so I think you are good to go.

What you could check is whether your slider bar (L ----- R) in the left box next to the soundwave window is centred in the middle. Sometimes that may be the cause for the DC bias.

But: this test is OK now and you are ready to roll !!! I hope you have fun with your first project. 8-)

thanks

Sonia

nathanielgillespie
Posts: 7
Joined: May 21st, 2019, 9:54 pm
Location: California

Post by nathanielgillespie » March 26th, 2020, 1:38 pm

Thank you so much for helping me!

So the slider bar shouldn't be in the center?

Kitty
LibriVox Admin Team
Posts: 22008
Joined: March 28th, 2014, 5:57 am

Post by Kitty » March 27th, 2020, 2:00 am

nathanielgillespie wrote:
March 26th, 2020, 1:38 pm
So the slider bar shouldn't be in the center?
sorry for being imprecise. The slider bar should be in the middle. But sometimes it isn't and that caused the extreme DC bias. Could be something else though in this case, I only tell you what has happened before to try to find the source of the problem.

Sonia

Post Reply