Ever since you first introduced this (v1 at the time), I started recording (audio books mostly) sitting on my chair in our living room, the TV or radio on lightly in the background, neighborhood traffic outside, the kids going back and forth and even the neighbor's dogs barking once in a while, and... EVERY single time, with NO processing, my recordings pass ALL ACX specs!
I tested this tool online. I thought the results were really good considering how much background noise it removes. Only my voice did sound a bit robotic with speech enhacement, and because I only had the free version, I couldn't adjust the strength. It's really good though, and I'm impressed.
For me i noticed it made my thin sounding webcam audio into a really nice podcast mix - with a lot more emphasis on bass which previously wasnt really there!
OMG - Amazing! Some wag came back to me with a piece of video audio with interviewer loud and the interviewee super quiet. Obviously, if you turn it up to hear the interviewee, you get LOUD INTERVIEWER and hissy interviewee. But I uploaded it like you suggested and - wow! Saved me so much time - and you can't even tell.. I've already liked this video - can I like it again?
don't worry you can! there's a little tab that says "v2" where you drag and drop audio. if you click that it'll open a drop down menu where you can select "v1". I stick to v1 myself, as it renders two speakers better
@@monojitpaul5483 You can either find 'links' where Adobe does not force you to pay stupid money, specially on subscription terms, or there are plenty of alternative good programs, either much cheaper, or a once payment ones. There are also good free programs you can live with.
Doesn't work.... "There was a problem connecting to Adobe Podcast. Try clearing your browser cookies and signing in again. If this problem continues, ensure that no browser extension or VPN is interfering with the connection."
For those who aren't premium members, we don't have the strength slider available. From the sound of a test file, I'm guessing the strength is set to 100%.
I've been playing with it recently and when I uploaded a file the default was 90%. However, the difference between 0 and 1% is a major improvement in and of itself. V2 is amazing.
@@michaelvalente4885 Thanks for the testing, Michael. That's great to know. I let my subscription run out and tried to go with perpetually licensed programs, but I really miss the Adobe stuff and this feature alone is making me want to reactive it, especially with the Black Friday deals going on right now.
I couldn't believe the difference. I had to use it on audio where the mic somehow got muted so I picked up the voice from another mic. Between 0 and 1 and it sounded like it came from the original mic. Amazing. I just wish I knew the process of how Adobe improves the audio.
Watched a free ai audacity noise removal that you did 9 months ago, are there any updates for free and local audio enhancements? Thanks a lot! Subscribed ❤
Thanks for sharing this, Mike! I'm going to go test this tomorrow (Monday). There's a tattoo shop right next door to my computer repair shop (where I'm also set up for recording my content) and even when their music is not very loud, it still bleeds through the sheetrock wall between our two units, and even with various tips you've shown on your channel, I can't get a clean, usable quality recording, for videos I record where I am (or at least my voice is) on camera. I didn't like the (mild, but noticeable) "electronic warble/flutter" that Enhance Speech v1 added to my voice, but this v2 model really seems to be up at a whole new level of quality. Can you imagine what they might be able to do by the time they have v3...?!? I can see them potentially adding some basic EQ or other adjustments to this tool, or even sound effects right into the tool, so if we DO want to add a bit more bass, or sparkle, or other tonal qualities to our recording, that they could add sliders or switches to be able to do that.
I actually just did a test at home using a Microsoft LifeCam HD (1920x1080/60) and a comica CVM-D03 clip-on lav mic that I got on sale for $36 last year (2 mics, the input box, & a 3.5mm patch cable into my PC mic jack). With v2, even on 11% or below, the difference is amazing! Even though the clip I used is only 46sec, no discernable difference in video quality going from the MKV file out of OBS, converting to 1080p/60 MP4 in Filmora, then uploading/downloading from Adobe Voice Enhance v2 for the 3rd iteration of the video file.
I've done more testing, and it seems like the high end gets reduced, which makes the sound a little "mushy," but it's easily fixed with a little EQ boost to the high end. Overall, I think it's a great tool for videos recorded outside that need a quick fix.
Okay, so here’s the big question: Is the enhanced speech of a 1% enhanced file identical to the enhanced speech of a 100% enhanced file? If so, you should be able to invert the 100% file, play the two files together, and hear ONLY the background noise. Can anyone with the premium plan confirm?
I say this is Very amazing if you are either having a Bad Mic, or trying to Recover Old Video audio, but it is kinda sad its not a Live thing, what i mean is instead of recording, we can just say it in the Machine Live Action, so when we live stream, the Microphone audio would go along the way, especially on Phone mics as well
This is just amazing! However, I have an audio file where the recorded audio from my guest's monitor was accidentally switched on. The echo is awful. V2 did nothing for that. Do you have any suggestions? I don't see where I can share the file here.
Hi Mike, Can you give me advice about podcast recording with MacBook Air m2 microphone. I have Au 2025. And I can use this Adobe enhance. What should I do respectively in Au? Thank you in advance :)
I'd recommend using the MacBook Air M2's built-in mic with a pop filter and adjusting the gain in Au 2025 to around -20 to -15 dB, then applying the Adobe Enhance feature to reduce noise and improve overall sound quality.
Awesome as usual Mike. I have a number of podcasts coming to 2025 so this is very timely. Quick gear question... I saw your vid about mics etc a few weeks back, but I would like to know what you are plugging your XLR mic into, and how that 'box' is connected to your computer.
Glad to hear my video was timely for your upcoming podcasting projects, and I'll do a follow up video on my audio setup to answer your question in more detail!
Have Adobe put this same quality version into Enhance speech within Premiere Pro. I and others have found quite a difference between using the web browser version compared to the Premiere Pro Enhance speech which was NOT as good
@@MikeRussell Thanks. So my question is, is this actually noise reduction tech, or is it cloning the speech via voice to voice and spitting it back out in a clean file?
Thanks for the video, Mike. I'm thinking of switching back to Adobe for video editing and wondered how best to integrate this tool into an Audition/Premiere Pro workflow for voiceovers. Would you suggest enhancing first and then finishing up in Audition or going directly into Premiere Pro and adjusting levels/compression there? Thanks!
Great question! I usually recommend enhancing your audio in Audition first for the best quality, then importing it into Premiere Pro for final adjustments. It gives you a lot of flexibility and control over your voiceovers!
It is pretty incredible, but you should know that from my experience with Hebrew audio (and I've seen comments from others about other languages), it doesn't work well with langauges different from English. I've seen a comment from someone who said that it's not really a sound enhancer, but a "speech reconstructor", and when you try it in different languages that seems like a fair distinction.
So you're saying it’s more of a “speech magician” than a “sound enhancer”? I guess it’s good at pulling tricks with English but struggles with the rest of the world!
@@MikeRussell Actually, when you understand that it's based on language and not just audio, it could give us a gliimpse into how and why it works so well: 1. It transcribes the text being spoken 2. It makes a model of the voice 3. It makes a new recording of the "performance", trying its best to mimic the tone of the speaker. If that's really the process, it could explain why the recording is so clean - there's no noise to begin with. It could also explain how it manages to clean the distortion so well, even though I guess it would be harder to make believable model of the voice. To be fair, we don't really know how the original voice sounds since we heard only the distorted version.
They use voice-clone tech to output a clean audio. I use Waves plugins to enhance my audio files and now a voice model of mine uploaded to Kits AI. It does a much better job than Adobe. Kits AI is the best thing ever invented.
is this an stand alone product or is part of another adobe product like premier pro? i cannot find enhanced speech on adobe product list. and if it is part of another product how is it accessed and activated? i even searched for adobe podcast and i dont see that as an App offered on adobe website. can you provide a link to the podcast/speech enhancer product
Mic bleed - your example enhanced both voices so I wouldn't say it's fixed because now you have two tracks with two voices very loud with slight delays - unless you use like phase alignment it's still going to sound bad (but now in a new way)
@@MikeRussell AS IF you actually respond to EVERY single comment despite having a channel with hundreds of thousands of subscribers! That's bananas and I hadn't noticed until you replied! Good on you! I wish I knew audio like you do, but I've definitely learned a lot from your videos so a massive thank you for everything you do!
The main problems is that it does work well in English, but when it comes to others languages, is less than ideal. Sorry.... but no! It's a bit un-natural tone, even on your voice. BTW.. I've tried multiple listing devices.. good headphones that I know (DT770pro, 7506 from Sony or M40x from Audiotehnica and monitors like 1029, 1030 from Genelec, Macbook pro m1 speakers... and so on). It's a good denoise plugin/software, but it's far away from studio like. I'll stick to other plugins/soft.. AI is good but don't over do it...
I appreciate your feedback! It’s true that language nuances can be a challenge, and I’m always looking for ways to improve the overall experience. Thanks for sharing your insights!
Naahh Version 1 is still better .. V2 filters out way to much Ambiente noice .. even at 1% its allmost gone .. sometimes i still want some ambient noice .. i hope adoub will let V1 to choos like they have at the moment .. V2 is to much ...
@@MikeRussell I also agree that V1 is much better because V2 kills all of your ambience and the slider isn't linear. The difference between 1% and 100% is so minor it's ridiculous. Lots of people are saying the same thing but Adobe insist this is a big upgrade on V1...
Ever since you first introduced this (v1 at the time), I started recording (audio books mostly) sitting on my chair in our living room, the TV or radio on lightly in the background, neighborhood traffic outside, the kids going back and forth and even the neighbor's dogs barking once in a while, and... EVERY single time, with NO processing, my recordings pass ALL ACX specs!
That's super cool! Love that you’re making it work with all that background noise. Keep rocking those recordings!
This is HUGE!!
Good stuff, Mike. I've been using podcast v1 for about a year and it's been awesome.
BUT you're right - it often adds bass and other EQ I didn't need.
Yep, v1 was all about that bass.
V2 is the same except now you have no control of the blending. 1% sounds like 100% 😮 V2 is unusable :(
I tested this tool online. I thought the results were really good considering how much background noise it removes. Only my voice did sound a bit robotic with speech enhacement, and because I only had the free version, I couldn't adjust the strength. It's really good though, and I'm impressed.
Good to hear!
For me i noticed it made my thin sounding webcam audio into a really nice podcast mix - with a lot more emphasis on bass which previously wasnt really there!
Where did you get the option to change the version?😮
You can find the version option in the drop down menu although it may have gone and v2 now default.
OMG - Amazing! Some wag came back to me with a piece of video audio with interviewer loud and the interviewee super quiet. Obviously, if you turn it up to hear the interviewee, you get LOUD INTERVIEWER and hissy interviewee. But I uploaded it like you suggested and - wow! Saved me so much time - and you can't even tell.. I've already liked this video - can I like it again?
Amazing! Yes, please do!
Great timing will start this today
That's awesome, can't wait to hear about your progress!
It's honestly so bad for me pls tell them to add a way to go back to v1... I don't need background noise removal too much
don't worry you can! there's a little tab that says "v2" where you drag and drop audio. if you click that it'll open a drop down menu where you can select "v1". I stick to v1 myself, as it renders two speakers better
@@RygarGuitarHey Rygar how? Because I've been looking for that so much at podcast audio enhancer
There's a drop down to change models.
I'm Starting to think that drop down is only available for the US So far because it's not available yet for us sadly...@@MikeRussell
@@MikeRussellI can clearly see the drop down on your video on mine its not there
I just renewed my "All Apps" Adobe subscription, I can't live with the audio editing alternatives!
Great!
You pay for Adobe stuff? Poor you.
@@motebike what should we use?
@@monojitpaul5483 You can either find 'links' where Adobe does not force you to pay stupid money, specially on subscription terms, or there are plenty of alternative good programs, either much cheaper, or a once payment ones. There are also good free programs you can live with.
Doesn't work.... "There was a problem connecting to Adobe Podcast. Try clearing your browser cookies and signing in again. If this problem continues, ensure that no browser extension or VPN is interfering with the connection."
Bummer! 😕 Those connection issues can be a pain. Just try clearing your cookies and see if that does the trick.
For those who aren't premium members, we don't have the strength slider available. From the sound of a test file, I'm guessing the strength is set to 100%.
I've been playing with it recently and when I uploaded a file the default was 90%. However, the difference between 0 and 1% is a major improvement in and of itself. V2 is amazing.
@@michaelvalente4885 Thanks for the testing, Michael. That's great to know. I let my subscription run out and tried to go with perpetually licensed programs, but I really miss the Adobe stuff and this feature alone is making me want to reactive it, especially with the Black Friday deals going on right now.
Ah, that strength slider IS helpful but I notice even 1% is quite a big change in v2.
I couldn't believe the difference. I had to use it on audio where the mic somehow got muted so I picked up the voice from another mic. Between 0 and 1 and it sounded like it came from the original mic. Amazing. I just wish I knew the process of how Adobe improves the audio.
@@MikeRussell I just renewed my subscription, and that slider definitely makes a huge difference. Thanks for showcasing this amazing feature!
Watched a free ai audacity noise removal that you did 9 months ago, are there any updates for free and local audio enhancements? Thanks a lot! Subscribed ❤
Thanks for sharing this, Mike!
I'm going to go test this tomorrow (Monday). There's a tattoo shop right next door to my computer repair shop (where I'm also set up for recording my content) and even when their music is not very loud, it still bleeds through the sheetrock wall between our two units, and even with various tips you've shown on your channel, I can't get a clean, usable quality recording, for videos I record where I am (or at least my voice is) on camera. I didn't like the (mild, but noticeable) "electronic warble/flutter" that Enhance Speech v1 added to my voice, but this v2 model really seems to be up at a whole new level of quality.
Can you imagine what they might be able to do by the time they have v3...?!? I can see them potentially adding some basic EQ or other adjustments to this tool, or even sound effects right into the tool, so if we DO want to add a bit more bass, or sparkle, or other tonal qualities to our recording, that they could add sliders or switches to be able to do that.
I actually just did a test at home using a Microsoft LifeCam HD (1920x1080/60) and a comica CVM-D03 clip-on lav mic that I got on sale for $36 last year (2 mics, the input box, & a 3.5mm patch cable into my PC mic jack). With v2, even on 11% or below, the difference is amazing! Even though the clip I used is only 46sec, no discernable difference in video quality going from the MKV file out of OBS, converting to 1080p/60 MP4 in Filmora, then uploading/downloading from Adobe Voice Enhance v2 for the 3rd iteration of the video file.
Great use case. Hope it goes well!
You used exactly the right word - INSANE!
Yep!
@@MikeRussell I just tested it standing next to a hair dryer. Just WOW.
Would you think shorter clips have better results since the audio changes less? or should I upload my whole audio of a 20 min out door walk & talk?
Up to 30 mins is allowed I believe.
I've done more testing, and it seems like the high end gets reduced, which makes the sound a little "mushy," but it's easily fixed with a little EQ boost to the high end. Overall, I think it's a great tool for videos recorded outside that need a quick fix.
Thanks for sharing your updated findings and providing a helpful tip for others who might experience the same issue!
Okay, so here’s the big question: Is the enhanced speech of a 1% enhanced file identical to the enhanced speech of a 100% enhanced file?
If so, you should be able to invert the 100% file, play the two files together, and hear ONLY the background noise.
Can anyone with the premium plan confirm?
Worthwhile question to ask!
I say this is Very amazing if you are either having a Bad Mic, or trying to Recover Old Video audio,
but it is kinda sad its not a Live thing, what i mean is instead of recording, we can just say it in the Machine Live Action, so when we live stream, the Microphone audio would go along the way, especially on Phone mics as well
That's a great point, and I can understand why you'd want a live solution for better audio quality during live streams!
@@MikeRussell true
The one in pr 2025 is still v1 as for now, right?
Yes, that's correct, the model available in PR for 2025 is still the first generation.
This is just amazing! However, I have an audio file where the recorded audio from my guest's monitor was accidentally switched on. The echo is awful. V2 did nothing for that. Do you have any suggestions? I don't see where I can share the file here.
Thanks for the love! That echo sounds like a real pain.
If i upload a 320 kbit file, the download is only 64 kbit. It sounds a little muffeld. I wold use it for radio production, if the quality was higher.
Shame to hear that.
@@MikeRussell Yeah would add quite a bit of messing around to my workflow. Shame as I think it sounds very good.
I have used Speech v2. I have found something interesting: The low frequency in some parts of a recording clip might be suddenly taken out!
Interesting!
Hi Mike,
Can you give me advice about podcast recording with MacBook Air m2 microphone. I have Au 2025. And I can use this Adobe enhance. What should I do respectively in Au? Thank you in advance :)
I'd recommend using the MacBook Air M2's built-in mic with a pop filter and adjusting the gain in Au 2025 to around -20 to -15 dB, then applying the Adobe Enhance feature to reduce noise and improve overall sound quality.
Awesome as usual Mike. I have a number of podcasts coming to 2025 so this is very timely. Quick gear question... I saw your vid about mics etc a few weeks back, but I would like to know what you are plugging your XLR mic into, and how that 'box' is connected to your computer.
Glad to hear my video was timely for your upcoming podcasting projects, and I'll do a follow up video on my audio setup to answer your question in more detail!
Have Adobe put this same quality version into Enhance speech within Premiere Pro. I and others have found quite a difference between using the web browser version compared to the Premiere Pro Enhance speech which was NOT as good
Right? It’s like they gave the web version a spa day while Premiere Pro is stuck in its pajamas! Let’s hope Adobe gets the memo!
Thanks for the video. Is this available in the Adobe suite or only online?
Only online right now!
@@MikeRussell Thanks. So my question is, is this actually noise reduction tech, or is it cloning the speech via voice to voice and spitting it back out in a clean file?
Thanks for the video, Mike. I'm thinking of switching back to Adobe for video editing and wondered how best to integrate this tool into an Audition/Premiere Pro workflow for voiceovers. Would you suggest enhancing first and then finishing up in Audition or going directly into Premiere Pro and adjusting levels/compression there? Thanks!
Great question! I usually recommend enhancing your audio in Audition first for the best quality, then importing it into Premiere Pro for final adjustments. It gives you a lot of flexibility and control over your voiceovers!
Download speed is extremely slow!!! Super frustrating! Anyone else struggling?
Sorry to hear that.
How does this compare to Resolve's AI audio noise reduction?
I think all AI noise reduction is improving daily. Adobe has traditionally done noise reduction best.
Please is this in Adobe audition software or this is done on a web browser?
Web browser but will likely be in audition & premiere pro very soon
@JoeCummings
Thanks for the reply. I love your videos.
Thank you! @@ROOMFORIMPROVEMENT88
No and I doubt it'll go into Adobe Audition but I may be wrong.
It is pretty incredible, but you should know that from my experience with Hebrew audio (and I've seen comments from others about other languages), it doesn't work well with langauges different from English. I've seen a comment from someone who said that it's not really a sound enhancer, but a "speech reconstructor", and when you try it in different languages that seems like a fair distinction.
So you're saying it’s more of a “speech magician” than a “sound enhancer”? I guess it’s good at pulling tricks with English but struggles with the rest of the world!
@@MikeRussell Actually, when you understand that it's based on language and not just audio, it could give us a gliimpse into how and why it works so well:
1. It transcribes the text being spoken
2. It makes a model of the voice
3. It makes a new recording of the "performance", trying its best to mimic the tone of the speaker.
If that's really the process, it could explain why the recording is so clean - there's no noise to begin with. It could also explain how it manages to clean the distortion so well, even though I guess it would be harder to make believable model of the voice. To be fair, we don't really know how the original voice sounds since we heard only the distorted version.
Yep. Kind of makes your voice robotic. I made a recording of myself saying I was testing it, and instead of "testing" it got "tested". A bit weird.
@@fzaatar Someone needs to look into this. Maybe talk to Adobe about it.
They use voice-clone tech to output a clean audio. I use Waves plugins to enhance my audio files and now a voice model of mine uploaded to Kits AI. It does a much better job than Adobe. Kits AI is the best thing ever invented.
is this an stand alone product or is part of another adobe product like premier pro? i cannot find enhanced speech on adobe product list. and if it is part of another product how is it accessed and activated? i even searched for adobe podcast and i dont see that as an App offered on adobe website. can you provide a link to the podcast/speech enhancer product
It's in the web browser only right now.
Does anyone know if v2 made it to Premiere Pro?
Never mind, answered my own question. Jason Levine mentioned it's not yet available in PP with no mention of an ETA.
@@bluejokenstewart Thanks, I'm sure we were all wondering. I was.
Not yet...
Mic bleed - your example enhanced both voices so I wouldn't say it's fixed because now you have two tracks with two voices very loud with slight delays - unless you use like phase alignment it's still going to sound bad (but now in a new way)
Totally get what you mean! It’s like trying to make a smooth smoothie with two stubborn fruits. Gotta love the chaos!
Game changer!
Yep!
Any alternative of adobe?
A few out there for sure!
we need an Enhance Speech for discord and other live voice chat apps, doing this only on recorded audio is not useful for streamers and gamers
I completely agree! It'll happen!
Sadly it doesn't work that well in other languages.
I made some tests and apparently many people agree.
Sad to hear this. Will only improve though!
Jaw drop!!!
Thanks!
It’s so bad that this is not done on-device
Maybe it will be soon!
Yeah this is pretty crazy. NGL.
Thanks!
@@MikeRussell AS IF you actually respond to EVERY single comment despite having a channel with hundreds of thousands of subscribers! That's bananas and I hadn't noticed until you replied! Good on you! I wish I knew audio like you do, but I've definitely learned a lot from your videos so a massive thank you for everything you do!
Enhance speech is that AI tool?
Yes it is!
For me it is not working properly.... previous version was better.....this version not giving better result
I found distortion removal incredible but depends what audio you use I guess?
30 minute max duration really kills it, most podcasts are over that.
Split files? I know it's not ideal.
sometimes sounds robotic
Thanks for the feedback, I'll work on making my delivery sound more natural!
The main problems is that it does work well in English, but when it comes to others languages, is less than ideal. Sorry.... but no! It's a bit un-natural tone, even on your voice. BTW.. I've tried multiple listing devices.. good headphones that I know (DT770pro, 7506 from Sony or M40x from Audiotehnica and monitors like 1029, 1030 from Genelec, Macbook pro m1 speakers... and so on). It's a good denoise plugin/software, but it's far away from studio like. I'll stick to other plugins/soft.. AI is good but don't over do it...
I appreciate your feedback! It’s true that language nuances can be a challenge, and I’m always looking for ways to improve the overall experience. Thanks for sharing your insights!
Naahh Version 1 is still better .. V2 filters out way to much Ambiente noice .. even at 1% its allmost gone .. sometimes i still want some ambient noice .. i hope adoub will let V1 to choos like they have at the moment .. V2 is to much ...
Interesting that you like v1 better!
@@MikeRussell I also agree that V1 is much better because V2 kills all of your ambience and the slider isn't linear. The difference between 1% and 100% is so minor it's ridiculous. Lots of people are saying the same thing but Adobe insist this is a big upgrade on V1...
tbh it seems be getting worse over time. The clip you posted sounds robotic.
Still a bit sometimes but not as bad.
Nah brother, you lied.. that's still feels like fake 🤥
Sorry to hear.
this not recognize different languages and make english become german …
Yes, it does use a model to recreate language!
This is nowhere near studio quality.
Practice makes progress!
how does it stay strong came up2..in
It sounds like you're curious about the concept!