Thank you very much for providing these informative videos. I currently use DxRevive, but I'm unsure why it only works sometimes. Could you please create a video about audio processing specifically for content creators? It would be helpful if the tutorial could cover technical priorities step by step. For example, when recording audio for a video, should we start by normalizing it or using plugins like DxRevive to improve it? Finally, how should we adjust the volume of the soundtrack? I often find myself confused about these steps. Which one should be the first to start?
Thanks for the good feedback. I really like your idea and will be working on a video about this. In response to your comment about dxRevive, there are a few things to consider. Not all noise and reverb are created equal. I have a collection of these tools because there are times when dxRevive doesn't perform well, but maybe Supertone Clear does. It really depends on how well our audio matches what the algorithm was trained on. If we look at Adobe's Enhance Speech, I would guess it was trained on truly bad audio because it struggles with good recordings that don't need a lot of work. This would explain why it can perform better than others at garbage audio, and struggle with something that needs a lighter touch. You can watch this video to get an idea of how there isn't one tool that always performs the best. ua-cam.com/video/r0kkwyZIey4/v-deo.html If you're using dxRevive on your own audio which is recorded in the same space each time, I would look into how consistent your recordings are from one to the next. If you're not setting everything up the same exact way each time you'll likely get varying results. The biggest key to good sound is to have to rely as little on tools like dxRevive as possible. Which problem is your biggest, reverb, noise, or are they equally problematic?
@@jesse.mccune Thanks for explaining the limitations of DxRevive and other plugins! That makes a lot of sense as to why Adobe Enhance Speech wasn't working as well for me. My biggest issue is definitely the reverb. I'm recording in a quiet room, but it's untreated and has a lot of echo. I've tried various plugins, including Enhance Speech, but haven't gotten the results I was hoping for.
Reverb can be the toughest element to deal with when working with dialog. If none of the tools you've tried have provided the results you're looking for, this suggests to me that there is likely a lot of reverb. You'll want to address this before recording rather than after. Lav mics are great for convenience and a clean look, but they are condenser mics and don't perform well in untreated spaces. For the average viewer, your audio is fine. The quality isn't such that people will stop listening because of the audio. I think that, given your current audio, putting a little effort into treating your space should make a difference. In case you haven't watched my video about the Yeti mic, I provide some examples that show the difference between recording in a completely untreated space and a treated space. It also shows the difference between condenser and dynamic mics in an untreated space. You can watch that video here. ua-cam.com/video/8h2LUw4_c-0/v-deo.html
You did an excellent job on this review. Clearly, this isn't anywhere near ready for prime time. I suspect management pushed developers to get an "AI" tool out as a minimal viable product (MVP) with the intention of better performance in future versions. I suppose this could be considered a MVP, but unless they improve it quickly it's not going to be widely used.
Thanks, Stephen. You nailed it, this feels like an MVP product. I question the logic behind putting out an MVP of something like this because there is already a lot of competition in this space and that first impression can really make or break a software release. I look at dxRevive as a really good example of how to release an MVP type plugin. It was pretty good out of the box. It was useable, but the EQ was too much in most cases, so I passed on it upon release while paying attention to how they improved it. Over the year or so since its release, they have added new algorithms and it's become my go-to tool for anything more than minimal noise and reverb reduction. Their continuous release of new algorithms really stands out because I'm not seeing this with the competition. It has a choice of 7 algorithms now so we can choose the one that works best for our material. Another good example of an MVP is the new Dialog Isolate in RX11. It performs well, but is poorly optimized, so we can only run a couple of instances before we run into overloads. I have no doubts that future versions on RX will see better optimization and tuning of the algorithm. Unless I start seeing a lot of buzz around it, I can't see myself revisiting Prime Vocal.
It could be if it performed well, but in terms of cleaning up dialog, Auphonic blows this out of the water. What functions in Auphonic are you looking to replace?
Thank you!
Glad you found it helpful.
Thank you very much for providing these informative videos. I currently use DxRevive, but I'm unsure why it only works sometimes. Could you please create a video about audio processing specifically for content creators? It would be helpful if the tutorial could cover technical priorities step by step. For example, when recording audio for a video, should we start by normalizing it or using plugins like DxRevive to improve it? Finally, how should we adjust the volume of the soundtrack? I often find myself confused about these steps. Which one should be the first to start?
Thanks for the good feedback. I really like your idea and will be working on a video about this.
In response to your comment about dxRevive, there are a few things to consider. Not all noise and reverb are created equal. I have a collection of these tools because there are times when dxRevive doesn't perform well, but maybe Supertone Clear does. It really depends on how well our audio matches what the algorithm was trained on. If we look at Adobe's Enhance Speech, I would guess it was trained on truly bad audio because it struggles with good recordings that don't need a lot of work. This would explain why it can perform better than others at garbage audio, and struggle with something that needs a lighter touch. You can watch this video to get an idea of how there isn't one tool that always performs the best. ua-cam.com/video/r0kkwyZIey4/v-deo.html
If you're using dxRevive on your own audio which is recorded in the same space each time, I would look into how consistent your recordings are from one to the next. If you're not setting everything up the same exact way each time you'll likely get varying results. The biggest key to good sound is to have to rely as little on tools like dxRevive as possible. Which problem is your biggest, reverb, noise, or are they equally problematic?
@@jesse.mccune
Thanks for explaining the limitations of DxRevive and other plugins! That makes a lot of sense as to why Adobe Enhance Speech wasn't working as well for me. My biggest issue is definitely the reverb.
I'm recording in a quiet room, but it's untreated and has a lot of echo. I've tried various plugins, including Enhance Speech, but haven't gotten the results I was hoping for.
Reverb can be the toughest element to deal with when working with dialog. If none of the tools you've tried have provided the results you're looking for, this suggests to me that there is likely a lot of reverb. You'll want to address this before recording rather than after. Lav mics are great for convenience and a clean look, but they are condenser mics and don't perform well in untreated spaces. For the average viewer, your audio is fine. The quality isn't such that people will stop listening because of the audio. I think that, given your current audio, putting a little effort into treating your space should make a difference. In case you haven't watched my video about the Yeti mic, I provide some examples that show the difference between recording in a completely untreated space and a treated space. It also shows the difference between condenser and dynamic mics in an untreated space. You can watch that video here. ua-cam.com/video/8h2LUw4_c-0/v-deo.html
You did an excellent job on this review. Clearly, this isn't anywhere near ready for prime time. I suspect management pushed developers to get an "AI" tool out as a minimal viable product (MVP) with the intention of better performance in future versions. I suppose this could be considered a MVP, but unless they improve it quickly it's not going to be widely used.
Thanks, Stephen. You nailed it, this feels like an MVP product. I question the logic behind putting out an MVP of something like this because there is already a lot of competition in this space and that first impression can really make or break a software release.
I look at dxRevive as a really good example of how to release an MVP type plugin. It was pretty good out of the box. It was useable, but the EQ was too much in most cases, so I passed on it upon release while paying attention to how they improved it. Over the year or so since its release, they have added new algorithms and it's become my go-to tool for anything more than minimal noise and reverb reduction. Their continuous release of new algorithms really stands out because I'm not seeing this with the competition. It has a choice of 7 algorithms now so we can choose the one that works best for our material.
Another good example of an MVP is the new Dialog Isolate in RX11. It performs well, but is poorly optimized, so we can only run a couple of instances before we run into overloads. I have no doubts that future versions on RX will see better optimization and tuning of the algorithm.
Unless I start seeing a lot of buzz around it, I can't see myself revisiting Prime Vocal.
Would you say it is a replacement for Auphonic?
It could be if it performed well, but in terms of cleaning up dialog, Auphonic blows this out of the water.
What functions in Auphonic are you looking to replace?