Audio Awful Post-Render

S

SDub

Well-known member
************RESOLVED******************

Hey Guys,

I picked up the Adobe trial to see if I wanted to switch from Vegas and am having a few issues. One issue I can't seem to get fixed is getting my audio settings/encoding settings correct. I have a video with three audio tracks. The three tracks are as follows (determined by the media settings in Premiere Pro CC 2014) :

1. 48 khz 16-bit mono (recorded Narrative to Tascam DR-40, cleaned and sweetened in audacity, exported to lossless .wav)
2. 16 kHz 16-bit mono (in-camera audio from DMC-LX5)
3. 44.1 kHz - compressed - Stereo (musical track, downloaded from some site)

When I render out the video, every single audio track sounds just as it does in premiere's video preview except for track #1 (which is the most important!). It sounds super thin and has an echo characteristic to it. I've tried all kinds of different settings including rendering all three tracks together as lossless PCM in Premiere and putting the track back in Premiere and the same thing happens (what!?). What's going on? Here are some pictures to help you help me!

Export Settings: http://imgur.com/CSsmqv4
Sequence Settings: http://imgur.com/fTknYAi
Track Properties: http://imgur.com/WBkEVyV
 
Last edited:
SDub said:
...except for track #1 (which is the most important!). It sounds super thin and has an echo characteristic to it.
Click to expand...

Without audio to listen to, this is just a shot in the dark. Literally. But... "super thin" can happen when two tracks containing the same or similar information are out of phase. Similarly, an "echo characteristic" can be caused by two tracks with the same or similar information being out of sync (alignment) by a frame or two. Both of these are typically caused by not turning off an audio track (camera scratch track for example) on the timeline.

But like I said, without some audio to listen to, it's just a shot in the dark.
 
Hi
Bruce above is the authority, not I, but just a thought that might help. Premiere CC 2014 makes some assumptions about what you are doing audio track wise. Read this Link: http://www.larryjordan.biz/adobe-premiere-pro-cc-separate-audio-channels/

Make sure the matrix of settings is to your liking. I am wondering if your mono track #1 is being treated as a stereo pair by Premiere. Again, just a thought. I think how this works (and doesn't...) is not always apparent to new users. Hope it may help. Good Luck.
 
Blit said:
Hi
Bruce above is the authority, not I, but just a thought that might help. Premiere CC 2014 makes some assumptions about what you are doing audio track wise. Read this Link: http://www.larryjordan.biz/adobe-premiere-pro-cc-separate-audio-channels/

Make sure the matrix of settings is to your liking. I am wondering if your mono track #1 is being treated as a stereo pair by Premiere. Again, just a thought. I think how this works (and doesn't...) is not always apparent to new users. Hope it may help. Good Luck.
Click to expand...

My track was being interpreted as mono, and it was recorded as mono, so we're good there. However, this made me revisit something I did in the past. I tried to export the audio as mono instead of stereo and it still sounds crappy...but a DIFFERENT kind of crappy. Here's an example: https://www.dropbox.com/s/8v28abpkjob2t19/Sequence 04_3.mp4

I'm not sure if this is by design but Premiere limits the max bit rate of audio to 160 kbps for mono tracks in the h.264 codec as opposed to 320 kbps for stereo tracks (twice as many channels). Is there a way to encode to a web friendly codec with a mono track that has a high bit rate?

Jim Simon said:
Try skipping Audacity and bringing the original recording from the Tascam.

Any difference?
Click to expand...

The above link is an example of using straight tascam audio. I tried the tascam audio using the same export settings as before, and had similar results.
 
SDub said:
OK, as promised here are some examples.

1. https://www.dropbox.com/s/fhubfkwv5nkoqt5/Audio Test 3.mp4
2. https://www.dropbox.com/s/36ipmm859gk2kg1/Audio Test 2.wav

#1 is an h.264 encoding w/no video, just audio for comparison's purpose. #2 is a lossless .wav encoding of the exact same audio track. For both tracks, I've specified the exact same settings for encoding other than the obvious different file type. Both were rendered with the settings as seen above in my first post.
Click to expand...

Interesting. My shot in the dark missed; no surprise there. I'm not hearing anything that's phasey, or that has an echo. What I'm hearing is a signal that's attenuated on both ends. Bottom end sound passable, but it sounds like it's rolling off the top starting somewhere in the presence region, say 4kHz or so. Severe compression could do this -- AT&T used to do this to audio over analog land lines, sounded a lot like this.

If you listen to this audio as you step through your workflow, and it sounds fine until you export, it would lead you to think that it's got something to do with exporting. But I don't know what. Or why.
 
Apologies for using the incorrect terms. I did use Audacity's parametric equalizer, but the example I posted in response to Blit's question was the encoding done on the track taken straight off of my Tascam DR-40. Here are my encoding settings: http://i.imgur.com/CSsmqv4.png
 
I see nothing wrong with the export settings. Unless you're a Golden Ears and have a system worth about $10K, the export should sound very nearly identical to the original.

Keeping the Audacity processing set aside and sticking with the original Tascam clip in PP for the moment, what exactly are you doing to that original in PP?
 
Literally nothing. I created a new sequence based on the DSLR 1080p format, made sure PP is interpreting the footage as mono (via hitting Ctrl+G while selecting the clip in the source manager), and rendering the first 10 seconds of the footage w/o video via the export settings mentioned in the OP.

To reiterate a question: Is there a way to encode the audio as mono, and have a higher bit rate (~320 kbps) while using a stream friendly codec like h.264 in PP?
 
Jim Simon said:
Well, that's not 'nothing'. It's also not necessary. Try your test without that step.
Click to expand...

By "checking to make sure it's being interpreted as mono" I just make sure that the track is mono. I don't change anything. It's set that way by default.
 
OK, I am stumped. I have no idea what to do. I recorded to my Tascam DR-40 in every single mode (Mono, Stereo, Dual, and tried w/the internal microphones on the tascam dr-40) rendered it in both Premiere and Sony Vegas and still, the audio for every single type of recording is not good.

Is this because of my recorder? The tracks sound lovely right off of the device but if the issue is consistent across two different NLEs so I have no what may be the culprit. I need to get this figured out because my trial ends on the 1st and I would like to have good audio for this current project.
 
Jim Simon said:
Try audio from another source. If it sounds fine after export, then it's looking more and more to be your Tascam.
Click to expand...

So, I went to Guitar Center and bought a brand new Tascam DR-40 and guess what?...the exact same thing happens. Wat. Additionally, I also recorded directly into Audacity w/my AT2020 mic, exported into PP, and still have bad audio distortion. I'm at a loss for words.

EDIT: I refreshed my PC, flashed my UEFI, installed all new driversm reinstalled premiere and AE and it works now. Must of been an interring 3rd party program, setting, or driver.
 
Last edited:
You must log in or register to reply here.
Back
Top