Street noise is something that’s very hard to deal with. My last place had a very busy street and any LP editing had a lot of noise removal involved There are things you can do, like making sure the isolation on your window frames is good, putting audio foam panels on your wall, putting in a divider that makes the L-shaped room small while you’re recording.
Without a recording room or studio, you’re always going to be dealing with some sort of editing or noise-gating while streaming.
Alright, cool. The divider and foam panels sound doable. My window frame are kind of garbage, so I guess I’ll be doing a bunch of retakes and noise gating. But I plan on doing post-comms anyway, so that’s relatively painless, just a fair bit of work if its busy out. Thanks!
What’s the viability of running an Avermedia U3 and playing off the preview window? i hear it has really good latency and I can get a u3 for like 60 quid right now. Anyone have any experience with this?
As a professional software developer, I’m quite content using Avisynth for all my video editing needs, thank you very much.
I agree that it’s not worth recommending to anyone who doesn’t at least program in C as a hobby, or hasn’t always wanted to learn, but I think the ability to do things like program frequently used edits and apply them with a single word, send those edits to someone in text, or download plugins and use whatever features they contain seamlessly, all taking up no more space than a text script and your raw inputs, is a pretty handy way to do your work if you can get your head around it.
Missed out on this but I’m checking out Hindenburg Journalist Pro - is anyone using this software? I’d love to hear thoughts on it before I throw down $375.
I fiddled around with the basic version last night, seems to be a marginally more user-friendly version of Audacity. Although unless I’m missing something super obvious (and I probably am), it doesn’t seem like it does Noise Reduction? Pretty much every other feature you’d expect is there.
I don’t know if there’s a specific way to make/load actual subtitle files in Sony Vegas (despite using it since Vegas 7, I haven’t dug very deep in to its options).
If I ever need subtitles for my videos I just go in to media generators and throw a text layer on.
From there you can treat it like a video clip. I specifically use the Legacy Text because oddly enough it seems to give you more control over your text (or at least I think so).
If you mean real subtitle files, like *.srt files or whatever, I just use Subtitle Workshop. Supports a comically large number of file formats and is generally very easy to use.
I recently got a new mic to replace my 5 year old Blue snowball, and I wanted to make a trip report on it because I think it’s a good starting point for people wanting to get into XLR mics but don’t have the budget to buy both a mic and audio interface.
The Audio-Technica ATR2100 is a cardioid dynamic microphone that connects with either USB (which is what I’m using now) or XLR (which I plan to use when I save up enough money for an interface. You could probably connect this with an XLR-to-USB cable and it would still work.) Being a dynamic microphone, I had to speak up a fair bit more than I’m used to to get it to about the same level of loudness as the snowball, but even then, that’s really not loud. So I don’t think I would use this in a live situation, but with a little elbow grease in Audacity I think managed to get the audio to acceptable quality.
And here’s the same sample after editing it in Audacity:
(I did drop the compression ratio from what I normally use which is about 4:1 to about 2:1.
I’ll need to play with it some more.)
Anyways, I think this is a pretty good mic and the option to connect via XLR opens up a upgrade path to something better down the road like a phantom-powered condenser mic.
Let me tell you about my favourite Audacity effect, called the Compressor. The compressor on paper decreases a sound’s “dynamic range”. What this means is that quiet sounds get louder but sounds that were already loud stay about the same. This effect is great for voice-over for a lot of reasons, but the one that stands out to me is that if you compress your voice right, you can then easily balance the volumes of your voice and other audio channels, like game audio.
My current workflow goes like this:
Import recorded voice and game channels separately and put the voice channel on “solo” to more easily work on it
Remove noise from the voice channel. This is best done as early as possible, because the compressor CAN amplify noise even with the right settings. Audacity comes with a noise removal tool that can eliminate radio- and AC line noise with the right profile, but for irregular noise I use the Noise Gate plugin, and sometimes a final, manual pass (ctrl+L is my friend for these) for any extra clicks that got through.
Normalize the track. The Amplify effect does almost the same thing if you just click on “OK”, but if you’re running the whole XLR shebang then Normalize can eliminate DC-offset as well, which could otherwise cause a “click” between audio.
Finally, use the Compressor. Try running default settings and see how that goes. If your track doesn’t feel “flat” enough, then lower the Ratio setting. If there’s some noise left and it gets amplified, try messing with the Threshold and Noise Floor settings. To make your messing-about faster, use a short part of your track to start with that you feel is a good example of the track as a whole, and then when you’re happy with your settings, apply those to the entire track.
Popping in to mention a minor interaction bug between OBS studio and Lightworks (for others just starting out, like me!): If you try to import (so far I have only tested MP4, not sure if other formats have this issue) a video file directly from OBS to LW that has multiple audio tracks, only the first track will be imported, but the video will still appear to have a second track. Except this second track (or with stereo, the second set of tracks) is just a duplicate of the first. Opening the video in VLC allows playing both audio tracks, so it’s some sort of software bug on the part of LW.
Fortunately, this is a pretty irrelevant bug because most if not all of the time you’ll be slapping the audio into audacity to fix it up, and afaik importing the audio file directly ignores this issue. I encountered this while trying to configure OBS for recording game audio and commentary on separate tracks for the local but not for the stream, so you can imagine my confusion when I imported the test video and didn’t have any game audio
Not sure if you’re aware, but the people who make Audio Hijack now also make a piece of software called Loopback, which, if I understand what Voicemeeter does, does pretty much the same thing that Voicemeeter does. It’s pricey, but in my experience, it works extremely well, with little to none of the setup hassle that a comparable Soundflower/Audio Hijack setup would give you. Worth a look.
Heads-up to anyone thinking of doing LP or streaming work on Linux: PipeWire is your friend. Whether you’re on Ubuntu or Manjaro, installing it is as simple as picking it in your package manager, and having that replace your PulseAudio package with the Pipewire-PulseAudio package. It will allow you to use PulseAudio-exclusive programs such as Discord with JACK programs such as Catarina.
…though you’ll have to install the PipeWire-Catarina package to do it. And unlike installing JACK, installing PipeWire will keep your PulseAudio and ALSA programs working just the way they did before. It’s taken streaming to Discord from a confusing nightmare to a matter of three clicks.
There are things you can do, like making sure the isolation on your window frames is good, putting audio foam panels on your wall, putting in a divider that makes the L-shaped room small while you’re recording.
