Audacity Forum

Hi
I am working with Vista. I have LAME. I can record and export as wav file without an issue. When I export as mp3 the result is a very tinny/echo-y sound like I am in a tunnel. This plays back poorly in Audacity and in other programs. The aup and wav files do not have this.

I would appreciate some help here. Thank you!

OK. That's a good one. The lame libraries are there, Audacity can open and use them, but they don't seem to work right. Fascinating.

Let's break the process in the middle and see which half fails. Download both of these sound clips, export them both as MP3, one at a time, and see what happens.

http://www.kozco.com/tech/soundtests.html

The idea is to process correctly-recorded sound clips that you didn't produce. Does the last segment of the Left-Right clip sound like everything you do, or is it the only one that sounds right?

Koz

There is one other possibility. Do you know what MP3s sound like if you encode them too small? You didn't use the magic word "bubbly," but that's what that sounds like. Check the encoding value in Audacity > Preferences > File Formats and make sure the encode rate is 48 or above.

Koz

Hello Koz
Thank you for your help.

I have exported both sound files into mp3 and the LF one has that echo-y hollow sound all the way through. It is not as noticeable on the piano one although it also sounds like it is less clear and sounds further away.

I can not find the word encode in properties file format. Is it called something else?

Obviously I am new at this but will eventually figure it out with some help! Thanks.

Norma

"Edit > Preferences > File Formats"
Bottom section "MP3 Export Setup"
Try setting the "Bit Rate" to 128

Was there ever any resolution on this problem? I have the same problem where my voiceover sounds undersea-ish only after exporting. It persists even after I have set the bitrate to 128. I believe I have the latest version of LAME installed as well.

bfuhr wrote:Was there ever any resolution on this problem? I have the same problem where my voiceover sounds undersea-ish only after exporting. It persists even after I have set the bitrate to 128. I believe I have the latest version of LAME installed as well.

Can you upload a sample somewhere - as you will probably have gathered from this thread, it's very difficult to identify the problem without hearing it.

Sure thing. http://podcasts.davenportlibrary.com/ep ... 030708.mp3

Sounds fantastic as a .wav or in .aup form. But once it mixes down the narrator vocals get buzzy, even at 128 kbps.

Thanks again.

I'd not say that the recording is bad, but it could be improved, so my comments here are made in the spirit of helping you to get better results and more satisfaction from your recording.

There's a few things that I notice straight away:
In the introduction at the beginning you've got a LOT of low bass frequencies - A bit of warmth to the voice is nice, but here my sub is making my trousers flap. There is also an underlying rumble coming from somewhere (possibly vibrations from your computer being picked up by the microphone?). MP3 encoding is optimised for music rather than voice, so it will try and preserve the highs and lows in the frequency spectrum, even though this may be detrimental to the intelligibility of your text.

Using a "spider" to hold the microphone can help reduce any rumbling vibrations coming up through the microphone stand. Slightly increasing your distance away from the microphone may help. Also a little bass roll-off applied (Equalisation) can help get the voice sounding clear.

When Joan starts talking there's some really heavy dynamic compression going on which is worse on the left channel (which has a higher average level) than the right. I'm unsure why this has occurred - possibly as simple as setting the recording level too high, or over-doing a "dynamic limiting" effect after recording.

It's good to aim for a reasonably high volume, but I think that you've over done it a bit. MP3 encoders frequently have trouble when the signal is banging up against the 0 dB limit, which is why I always recommend leaving a little headroom before encoding ( just 0.1 or 0.2 dB is sufficient). The problem is not limited to the encoder either - mp3 players will often perform poorly with clipped mp3 audio.

It is possible that these minor problems in the original recording have become more noticeable after encoding.

bfuhr wrote:Sure thing. http://podcasts.davenportlibrary.com/ep ... 030708.mp3

Sounds fantastic as a .wav or in .aup form. But once it mixes down the narrator vocals get buzzy, even at 128 kbps.

Thanks again.

Hello.. I am having the exact same issue. My recording sounds just like the one you posted here. What did you do to resolve the issue? Any help would be appreciated. Thanks..