Michael Hermes
Noise & Vibration Test Engineer
Podcast Producer
General Sound Guy
---
Podcast Producer/Editor Credits:
SoundCloud- Twitter
- Hermbot Audio
- Contact
-
Tracks
-
Tweets
- @chrisrking75 I was a believer! Thanks for a fantastic game, it's one of my favorites.
- @joshschultz @ejbenjamin Try telling them that trickle-down economics don't work and that polo is for jerks, they should leave.
- @RobZacny if we mention the 4x board game Eclipse we can call this episode "Total Eclipse of the Heart (of Darkness)"
- I love Necco Wafers.
- My new sounds: Crappy Audio http://t.co/DaGQJv0JMF on #SoundCloud
- @ArcenGames Hmm, what's Exodus of the Machine, a separate title? Thought I saw a forum post about two expansions this year...
- @AI_War @ArcenGames are there TWO incoming AI war expansions?
- @PAPPYBALL @jlhermes dill of course, what am I, a Scot?
- @PAPPYBALL @jlhermes we made those pickles
- @ejbenjamin in repose http://t.co/iGcTMD6JN5
- OH: "What handles stress better than steel?" "Adamantium." "Mithril." "An adamantium-mithril alloy."
- @AliSwoop @joshschultz Your certificate has been mailed!
- @joshjs a blender?
- Beautiful day for a ball game, let's play two! http://t.co/inXh8t1crS
- @PAPPYBALL am i doing this right ??
-
Posts
- Oct 24, 12
While I’m always tweaking and refining my podcast process, I wanted to take a moment and talk about the plugins I’m currently using. They’re created by Variety of Sound and are available for free download. I have tried many, many different free plugins and researched several paid options, but I always come back to the Variety of Sound options.
Every channel gets its own instance of preFix, which contains one of the better noise gates that I’ve found. I have never liked Audition’s gate, or at least I never quite figured out how to get the settings to a state that I was happy with. I like a longer release time, especially for noisier tracks – having the gate constantly drop in and out is more distracting to me than the noise. So while preFix has a lot of power and options, I only use it for the gate. The shelving EQ and phase options are nice, but not needed at this point.
After preFix, each channel gets its own NastyVCS (virtual console strip). This gem has everything that I’ve ever really needed: A solid parametric EQ section, low and high-pass filters, and a compressor. As with the preFix, each section can be switched on and off. Why add more plugins when this one does almost everything I need? I particularly like the compressor, which I leave set to the fastest attack and a medium release time. I’ve played with the tube saturation and limiter, but don’t use them on a regular basis.
After that, I only use a master bus limiter and EQ. The EQ is a high-pass filter set to cut everything below 32 Hz, which reduces any microphone rumble or other low-frequency nasties. Sometimes I’ll use the 20-band graphic EQ that comes with Audition if I have a particularly terrible sounding track, but that’s rare. If I’m really feeling froggy I might break out Spitfish, a very user-friendly de-esser. That doesn’t come up often, though.
In addition to being freely available and sounding great, the Variety of Sound plugins also get high marks for their beautiful UIs. I’ve said it before: presentation counts! If I could pay for these somehow, I would. Since the author doesn’t accept donations, however, I’m more than happy to write this up in hopes that someone will read it and head on over to the VoS download page.
- Apr 03, 12
This is the first post in a series that will examine the basics of audio production. Any one of these topics could be examined in depth over several articles, but I feel that painting in broad strokes is a good starting point. More technical discussion can follow.
What Noise Gates Do and Why They Make Life Better
A podcast is an excellent setting for demonstrating why a noise gate can be useful tool. As I mentioned in the previous post about podcast syncing, it can be useful to have separate tracks for individual contributors. A noise gate is a tool used to fix a common problem, specifically, background noise. We’ll start with the problem and follow up with how noise gates address it.
Three podcasters have recorded a show and sent you their individual files. After studiously syncing up the tracks and performing the final mixdown, all you hear is noise, noise, noise. Where did it all come from? Skimming through each of the tracks, the problems become readily apparent.
Podcaster A placed his mic right next to his high-powered, multi-fan gaming rig. His computer is cool, but his track is awash with fan noise. Podcaster B left his windows open, allowing the general background noise of the neighborhood to be included with his file. Podcaster C forgot to close his office door while his wife happened to be watching Law and Order: Special Victims Unit: Miami: Academy Squad. Put all of those noise sources together, and you have the recipe for a distracting cacophony of unpleasantness.
The only information we are interested in is the speech of each panelist. Everything in between is extraneous and distracting information. We could go through each track and mute all of the non-speech sections of the file. That’s tedious work, however, and podcast editors are a busy group of people – we have Ferraris to drive and yachts to polish. If only there was a tool that could get rid of all that background noise for us automatically!
Enter the Noise Gate. This is a tool that allows the user to set a volume level as a threshold. Anything below that threshold is gotten rid of. Everything above that threshold is left alone. With the proper setup, a noise gate can solve noise issues quickly and easily. Let’s look at an example:
Noisy!Here we can see the aforementioned fan noise, as well the speech we are trying to retain. I’ve inserted a blue line that will serve as our threshold. Note that it skates along just above the noise, but is well below the volume of speech. As I mentioned above, once we apply the effect, everything below the threshold will be removed, and everything above the threshold will be left alone. That means that the important part, the speech, will remain.
Here’s a picture of the same region after the effect has been applied. You can see that the noise has been removed, but the rest of the track is unaffected.
Much betterI only highlighted a small section of the track and applied the effect, so you can see at the left and right side where the noise comes back in. In between, the sections that only contained noise have been removed, and our noise problems have been solved.
Some of the more observant among you are thinking that the noise will still be present when the person is actually talking. This is true. However, the signal-to-noise ratio will be great enough that it will not be as noticeable to the listener, and that noise will not be compounded by the noise on every other track as we’ll be applying the same effect to everyone.
Who else can benefit from a noise gate? Guitarists are a good example. With high-gain amplifiers getting louder and louder, all of the background noise from the pickups to the amp are made louder as well. While the guitarist is playing it’s not an issue, but in between there is nothing to be heard but buzzes, hum, and crackling. A noise gate can clamp down all of that extra noise until the player starts up again.
Things to Watch For
One drawback to a noise gate is that, if set too aggressively, it can start impacting that content that you are trying to preserve. If you have a panelist who is a perpetual mumbler, for example, you might notice him or her trailing off at the ends of his sentences. His speech volume can dip precariously low to the threshold line I drew above, and his sentences will fade away to nothingness more than they already do. Because of this, I like to apply a noise gate after I’ve done any dynamic range adjustments such as compression or limiting.
Another note is that the utility of a noise gate diminishes drastically if used after mixdown. When all of files have been combined, a gate will only kick in when no one is talking. At all other times, the gate will be “off”, meaning all of the background noise is left in. As such, I’ll leave the noise gate as an online effect that gets applied at the same time as mixdown, meaning it’s the last thing to happen to each individual track before they are combined.
Conclusion and Further Reading
A noise gate is a useful tool that can solve noise headaches that plague amateur recordings. Most studios strive to be as quiet as possible to avoid having to solve such issues in the first place, but the real world is full of noise that is beyond our control. Gates provide a fast and easy way to eliminate or reduce background so that the focus can come back to the important information that we are trying to convey.
Noise Gates at Wikipedia
- Mar 31, 12
While some podcasts are recorded with the host and panelists all in the same room, many are recorded over large geographic distances. This results in several audio tracks that need to be time-aligned and mixed together in order to recreate the conversation.
One of the easiest methods to facilitate long-distance podcasting is to record a Skype conversation as a “master” track. This track has all of the audio from the podcast, but has several drawbacks:
- The audio is compressed, both in terms of bit rate and audio compression
- The audio can’t be edited on a per-panelist basis, which means coughs, background noises, and other noises that occur while someone else is talking cannot be edited out
- EQ, compression, or noise reduction cannot be performed for individual panelists
The Skype recording can work in a pinch, however, and many podcasts get by just putting up the Skype recording and editing out bits and pieces that aren’t needed. The alternative — the Hard Way — is to record the Skype track as well as a “local” recording for each panelist using a separate piece of software. Each local track will be of a higher quality than the Skype recording, but they will have to be mixed together.
Doing it the Hard Way is much more work, but the end product is worth it. Even the best content can be ignored due to poor presentation. (You’ll see me float this opinion a lot. It doesn’t matter if you’re cooking, making music, or editing a podcast. My mom always said: If it’s worth doing, it’s worth doing well.)
Getting in Time
The easiest way to sync up all the tracks is to have each panelist perform some easily discernible event at the beginning. Clapping or counting off numbers (or both) works, but the important part is that the event happens one person at a time. On our podcast we have everyone involved count down from 3 to 1 and then clap. It’s not 100% necessary, but it makes aligning their tracks very easy. The picture below illustrates that clapping offers a good visual indicator of how the tracks can be lined up.
Keeping Time
Is that all? Not quite. Podcasts can get lengthy and I regularly have to work with files that are an hour to an hour and a half long. Small differences in sample rate have compounding effects with such long files, meaning that the little changes add up to syncing problems later in the show. By the end of the podcast, unchecked sample difference can lead to a difference of seconds! Unacceptable. This leads to the panelists talking over each other and odd gaps in the conversation.
The good news is that it’s an easy and fast problem to correct. After you sync the initial slating cues, head to the tail end of the podcast and mute all of the tracks except for the first panelist and the reference Skype recording. Still synced? Maybe you got lucky and it is. Do the same with the next track – if you hear an echo or chorus-like effect, the track has lost its sync somewhere along the way. Listen to a few different places – has it gotten worse as the track goes on? Where does it become noticeable?
Find a place before it becomes apparent sync it up again. Place your cursor in a silent spot before the panelist starts talking, split the track, and realign the right portion to the Skype recording. Check again between the the split part and the end of the track. You may have to do it only once or twice, though if a track was upsampled by any significant amount you may have to do it a dozen times. The important thing is that the conversation still sounds natural and the responses from one person to another line up.
To summarize, here’s the basic workflow for syncing up the tracks:
- Import each panelist as a separate track.
- Import the Skype recording or other “master” track.
- Mute all tracks except for the reference recording and the first panelist.
- Align at the beginning of the track using whatever slating signal works best.
- Check the alignment at the end of the track.
- Split the track and nudge it, if necessary.
- Repeat for each other track until all of them are aligned.
- Mute the reference track.
- Mix down to final recording.
- Mar 06, 12
For Windows users, the first go-to free audio application is Audacity. Audacity is cross-platform and also available for OSX and your favorite flavor of Linux.
Audacity is a great solution for audio editing, touch-up work, recording voice-overs, and simple multi-tracking. While I’ve read that some bands have recorded full albums with it, that seems to be shoehorning the software into a role it was not designed to fill. The lack of MIDI support (which is expanding) is also a barrier for the music-recording space. As a general-purpose audio editing application, however, it’s a great tool have to installed.
The Good
Audacity is a powerful editor. A user that is interested in modifying, restoring, editing audio, or creating sound effects will find that Audacity is more than up to the task with a large pool of built-in tools.
Audacity is an easy to use recorder. After recommending the software to several friends and installing it myself on several different machines, Audacity does a fantastic job of “Start software – press giant ‘record’ button – get results”. It does a good job of detecting hardware and the incoming sound source, be it my external audio interface or a laptop microphone.
Audacity is still actively maintained. Some open source projects are abandoned after the developers lose interest or the community lacks direction or planning. This hasn’t happened to Audacity, which is still actively supported with bug fixes and new features. Support is easy to find with their online documentation, wiki, or forum.
The Bad
One criticism I would level is the lack of a built-in mp3 encoder. This limitation is borne out of philosophy and necessity, which I understand. Audacity, being Free and Open Source Software, cannot distribute proprietary libraries as part of its code base. As mp3 is the file format du jor, working with them is a common task for most users. Convenience counts, and a one-click installation would go a long way in making it more accessible to new users. Instructions for preparing MP3 support are available here. Similarly, I would like to see VST support out of the box (and a little easier to configure – the topic for a future pose), but that’s probably not as widely requested for as proper mp3 support.
The Ugly
I’ve watched the development of Audacity for years, eagerly awaiting the day they update the interface. Many will rightfully point out that the interface is functional and doesn’t affect how the software actually works. I agree. However, ask any chef: presentation counts. I’m a vocal proponent of the software and when people ask for a quick solution to edit an audio file I always point them to Audacity. First impressions are always that software is old and out of date because the interface is… old and out of date.
Again, while I have small criticiams against the bigger picture of the capabilities of the software, this is the conversation I have about 80% of the time after enthusiastically pointing someone toward the software:
Me: “Here’s Audacity!”
Friend: “Why does it look like it’s from 1990?”
Me: “Don’t worry about that, it’s a great piece of software.”
Friend: “If it’s so great, why can’t I open this mp3 file?”
Me: “Oh, uh, you just have to take these extra steps to make it do what you want.”It doesn’t instill confidence in a new user, and it puts the software at an immediate disadvantage. I must stress again, however, that these small complaints pale in comparison to the overall utility of the software.
In the near future I’ll write some user guides for touching up a lackluster live recording, restoring old audio, and recording basic voiceovers using Audacity.
- Jan 22, 12
I finally broke down and bought a full license for Reaper, which I am enjoying immensely. I also recently found GreenMachine 2, a VST guitar amp sim that sounds great. Here’s a track I threw together to highlight some high-gain fun:
- Aug 01, 11
I stumbled upon sfxr on the gamedev subreddit, a useful stop for anyone interested in game design. sfxr creates eerily accurate NES-style sound effects such as bleeps, bloops, and the occasional blorp.
As noted on the website, the easiest way to get started is to hit the “random” button until you stumble on something useful, after which you can tweak to your heart’s content. Sounds can be saved for later modification and exported as wav files.
There are starting points for the following sound categories:
- Pickup/coin
- Laser/shoot
- Explosion
- Powerup
- Hit/hurt
- Jump
- Blip/select
sfxr is free, open source, lightweight, and incredibly useful.
