Feedback on technical quality
Got my first try rejected because of some plosives and too much processing. Looking for some feedback on the technical sound of this clip before I get into resubmitting another Bunny script. Hopefully, it's got that non-popping, "natural" sound. Thanks! Mike
https://www.dropbox.com/s/ovwwacnvysw1shz/mikeberlak267755liver.mp3?dl=0
Please sign in to leave a comment.
Hi Mike!
It does sound quite filtered. It's very sibilant and includes several plosives. I can also pick up som background hiss which could be caused by the extra gain that's been applied. I suggest you lower the gain from your preamp if you're not compressing the recording!
Would you mind letting me know which equipment do you use? Also, can you share with us a raw recording so we can check it?
OK, Angela, thanks for the review. I want to get this right so I've done a reset of all settings and recorded some comparison clips for you with and without processing, noise reduction etc. The 80Hz lo-cut filter is not engaged through any of this.
Here's the setup: Sennheiser 416 with windscreen --> dbx286s preamp/processor --> Focusrite Scarlett Solo 2nd Gen --> Audacity.
Here's what you'll hear on this attempt:
1 raw with all dbx processing switched out. 123456 123456 WXYZ
2 here it is with my new settings for very light compression, all of 2 little clicks of high enhance to make up for a little loss when compressed, and 2:1 on the gate.
3 exact same file as #2 (processed) but also adding Audacity noise reduction set at 16/6/0.
NOTE: clips 1,2 & 3 were done when the room was totally quiet. There is a furnace/air conditioning unit in the attic right above my studio space. This is a given I've got to work with...no chance to move that or the studio :) Next are some clips with the furnace running.
4 room tone with furnace running in attic above me... no processing at all. (Yes, it's loud!)
5 room tone with furnace running but with just the Audacity noise reduction applied.
6 voice count 123456 with furnace running and no processing at all.
7 voice count 123456 with furnace running and just the Audacity noise reduction.
8 back-to-back comparison of #1 and #3 above...(totally raw vs totally processed, noise reduced) 123456 123456 WXYZ on each.
To be honest, I'm listening on pretty decent speakers and can't tell that much difference in quality. When the furnace is off, I do like the clean, non-processed sound for most things...with the processing only used as required for effect. When the furnace is on, I don't think we can get away without some form of noise reduction and I really don't hear any degradation to the voice when it is used.
What do you...and any others who might want to chime in...think?
THANKS!
Here are both .mp3 and .wav versions ...
https://www.dropbox.com/s/xp7pqywftkc3eu0/mic%20chain%20test%20MP3.mp3?dl=0
https://www.dropbox.com/s/mp80fjus3nw4xzp/mic%20chain%20test%20WAV.wav?dl=0
Hi Mike!
Thanks for sharing the files! Very informative and comparative. I really liked it!
So, listening to all of them, I liked the voice sound unprocessed and processed without noise reduction. For the processing one, though, I would bypass the compressor altogether. I believe the threshold may be a bit too low so it's basically compressing everything making it too tight. The sound with just the gate is very nice and I would add a high pass filter maybe around 60Hz - 80Hz to reduce some of the lower end. If your mic has a pad, you can try turning that on!
The noise reduction from Audacity is not of the highest quality so it does filter the recording quite a bit and changes the overall sound of the voice. This is when the recording becomes sibilant and when the breaths sound digital and filtered. I would definitely avoid using it and try to fix the hiss from its core.
When we listen to the room tone only, it definitely improves a lot when you add the noise reduction because it's not attempting to filter it in the background. When you have a voice on top, it makes a huge difference because the algorithm needs to be able to remove some of the noise's frequencies without damaging the voice too much. This is more challenging for the plugin which is why sometimes the take can sound very digital or metallic. Makes sense?
For the hiss, maybe it's coming from the space as it's so present in the recording. Have you tried changing the positioning of the mic in the space?
Your voice sound really, really nice! You have an amazing tone and the overall sound is really amazing. I say you should request the new application project now :)
2nd attempt:
Here are the issues found in your application:
Overcompression, you can hear it clearly throughout your recording.
Room echo, you can hear it clearly throughout your recording.
OK one more time.
Angela, I don't know if it's you or someone else who reviews the applications, but with one attempt left, can you let me know if THIS one would make it? ALL processing in both the dbx 286s AND Audacity are switched out...this is mic to preamp to interface only. If it's good, can you consider this my 3rd attempt? If it's not, let me know and I shall move on :)
https://www.dropbox.com/s/q3tcp43ij7mk4rn/bunnytest3.wav?dl=0
Hello Mark!
Listening to your audio I still hear the issues mentioned by the QC team. There is loud levels of hiss in your recording. You may want to try lowering the gain of your preamp and then raising the volume in the DAW through normalization to see if we can lower that noise. What equipment are you using?
Additionally, the space where you are recording still has many reflections and it creates that room echo effect. You’ll find tips on acoustic treatment on our Community page: https://goo.gl/vGqUIW .
Finally, the audio sounds heavily processed, what processing are you adding to the chain?
Best,
Juan
Hi Mark!
Thank you for sharing the recording.
Juan is our Head of Quality Control so I believe his comments are the most valuable! I agree with everything he said. I'd appreciate if you can let us know how you're recording including equipment and processing.
Hi Angela and Juan -- Thanks for the latest review.
The regular chain here is:
Sennheiser 416
dbx286s pre/proc
Focusrite Scarlett 2G
Audacity
I went thru everything last nite with headphones up real loud and found the following:
The 416 is a VERY sensitive mic! It was picking up the buzz of the power supply of the dbx, which was a couple feet away on the desk. AND it was hearing a high-pitched whine from the computer monitor which was right next to it. AND, it also was getting a real low-level hum coming from the speakers attached to my monitor amp. I have relocated all the offenders and I hope I've got a much cleaner sound. I would think all that junk must have been the "hiss" that you both reported.
I also have spread a nice comfy blanket over the desk top to hopefully soften reflections from there.
The one thing I can't get away from is the heating/ac unit in the attic right above me. This thing is on and off all day long and the only practical way to kill that noise is software noise reduction.
So.... Here is a new file I just recorded this morning...and this is just the mic directly to the Focusrite. The dbx isn't even plugged in to the chain at the moment. There IS noise reduction on this and a pass of iZotope "Mouth Declick."
What do you say?
Thanks,
Mike
https://www.dropbox.com/s/ixvzsig20tcaxyi/bunnytest0208.mp3?dl=0
Hi Mike!
Thank you for the explanation. The post processing you're adding is definitely damaging the recording a lot. The noise floor is too high and the recording is even distorting at some points which creates even more hiss. I suggest you remove the processing altogether as it is not helping!
Hi Mike!
I like how in the new recording, your voice sounds closer to the mic. It doesn't sound distant and with room echo. I still hear a constant noise in the background that's masking your mic. It's loud enough that the breaks sound weird, for example at the very end, the "every day" sounded to me as disconnected from the previous line "dream bigger...". I think that's because of the short gap of digital silence. The A/C and all other appliances like refrigerators, computers and such make a lot of noise that our brain is able to block from our attention. It'd be a great experiment to record 10 secs of room tone (just the room in silence) and see how loud it is.
Are you adding EQ or compression of any sort besides the noise reduction plugin?
Best,
Juan.
Juan/Angela --
Thanks again to both of you. No, there was NO EQ or compression on that previous sample. Just the Audacity NR and the iZotope Mouth De-Click.
SO..... the science project continues..here's another new one for you :)
(I have also added a bit more sound treatment...this time to the ceiling.)
https://www.dropbox.com/s/7l8bs3221yv604b/bunnytest0213.mp3?dl=0
This one starts out with about 5 sec of room tone (which looks to be at about -60db).
THEN you have two reads of the same commercial tag, one 7 inches from the mic, and one 10 inches away...noticeable difference! (Not sure which is really better, but there is a difference.)
There is NO NOTHING on this sequence...no lo-cut filter, no compression, EQ, noise reduction, declicking...it's just Sennheiser 416 to the preamp to the interface. Levels look fine on the voice: -6, occasional peaks at -3. There should be no distortion. And while this was being recorded, the heat/AC unit above me was off. So any "noise" picked up by the mic should just be the quiet ambience of the house!
I then copied and pasted the above "raw" sequence THREE more times and applied different noise reduction plug-ins to each of the three for comparison purposes...
Looking forward to your reports on this attempt. Hey, this is a $1,000 mic so I was hoping for a million dollar sound :)
Mike
Hi Mike!
Let me go over each part of the file to make it easier to explain. However, in general I can tell the volume is too high overall. The first thing I'd say, would be to tone the gain down. Also, the edits in general are too harsh. The beginning sounds super hard and a bit clippy so I suggest you be careful with how you edit and fade the clips.
1. Room tone: The hiss is very loud and I'm sure it's coming from something in your recording chain. From the chain you shared, I would take the preamp out and try recording with the 416 directly into the Scarlett. It could not be that, but it's worth a shot just to see if it could be cause.
2. 0:05 - 0:28: This mic definitely gives you an amazing vocal sound very rich. However, overall I can tell the sound is a bit sibilant. This could be because you're recording with too much gain and/or too near to the mic's capsule. The first take, when you're at 7 in, the voice sounds too upfront. This sounds as if you were recording too near to the mic so I would say it's better to go with 10 in. However, this still sounds too upfront. I believe this is caused by the excess gain, too.
3. 0:32 - 0:56: The voice sounds quite compressed, especially on the first take. I believe this is caused by the processing you're adding. Denoise plugins are a combination of processes so I'm sure it's adding some compression there. I don't really feel the difference with the noise thoug.
4. Last sections: These last sections are, to me, a no go. The processing is definitely adding a lot of filtering that is damaging the response an level of the voice too much. I'd definitely toss them away and take care of the noise without any plugins.
Keep in mind that the microphone is not the only part that's important to take care of when you're recording. The interface, preamps, cables and even the computer ports are vital to achieve a clean recording. You can have a $10,000 mic but it won't be able to bring it's full potential if it's connected to a $50 interface! Hope that makes sense!
Angela... Thanks. Just curious, what are you listening to these files on at your end? I'm using Sony 7506 headphones here plugged right into the Focusrite interface and I hear absolutely no hiss. The Focusrite is acting as the computer's sound card so the "regular" cheap sound card is bypassed, and I'm not listening thru any amplifier, just the headphone out of the Focusrite.
I'll get you a sample soon of mic direct to the interface and see if that's any different.
Also, I do notice, especially with headphones, a decent amount of mouth crackle on the files of 0213. There was no attempt to get rid of that, although I do have the iZotope "Mouth De-Click" which works very well. Maybe that is the "hiss" you are talking about. I do have a noisy mouth!
Hey Mike!
I think improving the acoustics of your room will make a huge difference. There's still a noticeable room echo, that's affecting the clarity of your voice. Additionally, as you mention, there are numerous mouth clicks that distract form the performance. To correct this, you may want to re position your mic, and perhaps try recording off-axis to the capsule. That will help us minimize the mouth noises and plosives.
Watch out for mouse clicks and other background noises!
Best,
Juan.
Juan & Angela --
Here are a couple new files.
First is a RAW read of part of a commercial (direct from the mic to the Focusrite interface) followed by a few seconds of the room. There is absolutely nothing on this as far as hardware or software processing. It sounds very quiet as far as room tone or hiss (none heard at all) or anything else. Even after normalizing, the level of the roomtone is at -60db, certainly acceptable.
The other file is the same commercial done the "usual" way, thru the dbx preamp/proc with some cleanup processing from RX6.
Thanks. Curious: What speakers/amp/or headphones do you guys listen to these on?
Mike
https://www.dropbox.com/s/ktgfo1oqeze7bvf/bunnytest0226-RAW.wav?dl=0
https://www.dropbox.com/s/gs8azvhi37oeztn/bunnytest0226-PROCESSED.wav?dl=0
Hi Mike!
We're using the Sony MDR 7506. I also use the AudioTechnica ATH-M50x to get an extra monitoring perception.
Now, I tired to open the files, but Dropbox gives me a 404 error saying the file is not found. Can you please check that and share them again?
Trying the dropbox links again...don't know what was up with that...they worked yesterday :)
https://www.dropbox.com/s/9r2nv2ch01hqcbs/bunnytest0226-RAW.wav?dl=0
https://www.dropbox.com/s/tpijj97b85myqb2/bunnytest0226-PROCESSED.wav?dl=0
Hi Mike!
Thanks for sharing the links again. This definitely work now :) Not sure why they didn't work before!
So, I definitely, 100% like the raw take better. It sounds really nice! It just sounds a bit too upfront. Would you mind recording a bit farther away? Also, some of the edits sound a bit too harsh, however, I don't know if you made any cuts here because the recording sounds as if it wasn't touched. if this is the case, then the harshness I pick up is coming from recording a bit too near to the mic's capsule. Try backing off a bit :)
Also, I suggest you edit some of the mouth noises and breaths you have there. You can cut them and use fades to make the transitions smoother and I'm sure this will provide a cleaner, much nicer take.