Enhance Speech v2 is here!

[Harmony J]

We’re thrilled to introduce Adobe Podcast Enhance Speech v2 to our community. Our team has been hard at work to bring you a tool that transforms your audio and video recordings into studio-quality sound with just a few clicks.

Here are the v2 highlights:

Natural clarity: Clear, natural dialogue without robotic tones—no matter the language or recording conditions. It handles everything from noisy crowds to poor acoustics and delivers studio-quality audio.

Strength slider: Powered by our Sound Lift AI model, the updated slider allows you to balance speech clarity with ambient noise. Want a bit of café ambiance in your podcast? Adjust the slider to find the perfect mix.

Noise removal: Background noise, reverb, and music—v2 removes it all, keeping your message sharp and clear.

Clarity boost: Struggling with quiet or distant audio? V2 improves clarity in soft recordings without amplifying unwanted noise.

Try out Enhance Speech v2 here:

podcast.adobe.com/enhance

We’d love to hear your feedback and suggestions. Your thoughts are super valuable as we continue to improve and innovate.

We’re excited to see how you use v2 to elevate your audio storytelling!

[Lpi_Ral8804]

Even so, I think it would be better if there was an option to choose between V1 or V2 because not everyone necessarily likes the audio produced by V2. For example, in my case, I personally prefer V1 and feel more comfortable with the sound it produces. So why isn’t there an option to choose between V1 or V2? Why are we forced to use V2?

[Arsen_Yasu7211]

++++

[erics.simmons.401]

Adobe People... Enhance Speech v1 saved my butt on an expensive work project. For that, I will forever be grateful... I was hired to shoot a guided tour in a live compounding pharmacy, and the background industrial noise (that I couldn't control) was so bad that I honestly thought my career was over. Enhance Speech v1 cleaned the audio subtly to the point that it was usable. Nobody even knew what I'd done. Wonderful tech. What I'm hearing in v2 is so alien, obvious, and artificial that I could never utilize it. Yes, it's clean, but It no longer sounds like the person who was speaking. This is analogous to an AI photography sharpening tool that doesn't actually sharpen a person's portrait; it replaces their face with a totally new face and, therefore, becomes unusable. Please make v1 available again for those who want it. Thank you. 

[erics.simmons.401]

Wait... I was freaking out, but now I see that v1 is still accessible. Please keep it that way. THANKS!

[Arsen_Yasu7211]

How did you switch

[Someone_Sone2728]

I REALLY NEED TO USE V1 VERSION!!! Where can I find it?

[erics.simmons.401]

If you're signed in and go to: https://podcast.adobe.com/enhance, there is a downward arrow toggle (default is v2), allowing you to access v1. Since my post a few days ago, I've compared results using both versions, and I've gotta say that v1 keeps winning. In my opinion, it sounds natural and preserves enough of the audio's original character to not cross into uncanny valley territory. Adobe, keep experimenting with this tech, but PLEASE keep v1 available as a tool for those who prefer it. Thanks.

[Lonesome MT Hybrid_Mo5722]

There's no toggle for me sigh V1 was the best idk why Adobe likes doing this they don't even care enough to read comments 

[Paul38916060g8i9]

If you toggle the little button to v1 in the upload window you can use v1.

Once you upload you cannot switch it to v1.




Paul Salveson | Director of Audio

[cid:ramseynetwork-sig-logo_81a54c0c-045b-45b2-8cdc-96ff4750b7de.png]

Ramsey Solutions
1011 Reams Fleming Blvd., Franklin, TN 37064
O 615.850.1268
ramseysolutions.com<>

[Arsen_Yasu7211]

PLEASE LET ME USE V1 BACK

[Arsen_Yasu7211]

The sound in v2 does not allow me to do my job, it just kills my YouTube channel, I am depressed and standing still because I am afraid that people subscribed to me and watched the video only because of the voice, which sounds extremely terrible without processing, please i want v1 back... 

[jay_scott2337]

absolutely

[Henrik Heigl]

Hi,

I tested it with the following Test Scenario: I assume I am at a concert as a young journalist and want to interview someone. In the background the Band on the stage plays and its loud, much noise from the crowd, other voices in the background, etc. I use a smartphone as the recording device (in my case an iphone). To simulate that scenario I searched on YouTube a Live concert, play it over some Bluetooth speaker or something similar and use the smartphone as a microphone to record that staged/fictive interview with myself 😉

After that recording i simply upload that file into the enhance speech.

Results:

The Tools makes a pretty good job! V2 is definitely WAY better, no more "robot voice", etc. if it comes to making the voice better, but its not yet there. Especially if the background noise changes within the recording also the Voice-color changes and thats a bit odd. if the background is much louder then the own voice or if there is recognisable second voice that is talking in the background and you make a pause then also the background voice is recognized as the own voice.

Conclusion:

I would wish for some more "control knobs" e.g. enhance speech, background overlay intensity, voice color (more/less radio voice), more/less compression, more/less AI analysis, etc.

Also, separation of voices (Voice1, Voice2, background music, etc.) so that those can be edited later in the Adobe Podcast Studio separately. A Button "edit with Audition" and "edit with Studio (beta) is also needed.
Hope that helps. Also it maybe possible not only to download as wave, but also as mp3 or other formats?
With those additions I could imagine also some kind of "Enhance Speech VST" as Plugin for Adobe Audition or other DAW's (just an idea).

[CGBESSELLIEU]

Agreed, control knobs for various key parameters would make the tool more robust!

[NAF22]

This version is terrible! It filters out things it shouldn't, like laughter, and the percentage slider is totally useless now. 

[CGBESSELLIEU]

Unfortunately, can confirm it does filter out laughter.
And with the broken slider can't test different percentages on that.

I recognize its likely difficult to design these models to differentiate between noise and laughter.

[miguel_bisono7399]

Are those changes applied to adobe premiere rush?

[CGBESSELLIEU]

V2 sounds better quality, preserves more of the voice (rather than becoming too robotic/processed at high %s).
Reporting Bug: the slider is not working for me. 1%-100% all sound the same. 0% goes to original unprocessed sound.

[NAF22]

The slider not working is kind of a major issue, right???

[CGBESSELLIEU]

Yes, I'm surprised it shipped that way. It's still broken unfortunately... hope a patch is in the works!

[WisjonPhoto]

To the support/developers: Its great to boost the sound no argue there. However, it’s to heavy on the boost and the clean up. It filtrating away laughter and sometimes makes the vocals a bit "robotic".  As it is now I find it pretty much usless. My suggestion is that you separate background noises/echo and boost with separate percentage sliders. And learn the Ai to keep laugher in. I also noticed that if you "make the voice" that you normally don't make. Then it filters extremely strongly and everything sounds very strange. I thought this app would save hours of time from manual editing.

[CGBESSELLIEU]

@Harmony J curious, is the V2 model only built into Adobe Podcast on the web or is it also built into the current version of Premiere via Enhance Speech? Or would that still be V1 currently?

[Harmony J]

Currently, Enhance Speech v2 is only available in Adobe Podcast. Premiere is still using v1.

[Camille_M8726]

I use Enhance Speech for my podcast and it is a life saver. However, I agree with previous comments: 

- please LET US KEEP V1! 

- V2 looks very promising but filters too many "emotions" like laughter and surprises like "oh, wow"

- very strange and PLEASE look into this: it feels that now the sound is not constant anymore, meaning that at one point my guest is speaking super lound, then it goes low, then goes back up... very strange. 

- would love to see some discounts for long time users 🙂