Unmixr AI - Text to Speech, Speech to Text, Audio/Video Dubbing Reviews

If your goal is to upload pdf/documents/epubs, etc. to convert them into audio, you should know that the platform does not have that option. You have to manually copy/paste, which is not practical when what we need is to convert a long document, manual, book, etc.
I suggest the developers add this feature: convert pdf/documents/epubs into audio.
The TTS do sound natural and have high quality.

We're in the weird transition to TTS, and in 2-5 years this will all probably be free. But for the time being, bulk TTS is costly, and I have used DOZENS of these services (including many from Appsumo - Play.HT, Speechki, Unmixr, VoiSpark and Uttr which became Blakify) over the years.

This one is one of the better ones, and I'm tempted to get a 2nd full stack (if that's allowed - or if I can stack beyond the max stack, I'd do that). Why?

1) Most of these voices have "tells" that they're AI - but nearly all AI - even best-in-class AI - have their hiccups. But they have ways to let you modulate many of them to make them far more malleable than other platforms.

2) This lets me do hour-long podcast episodes. They call it the "Long Voice Studio (V2)" currently. A "full stack" account currently gives you 1.1M characters. One hour long script is about 65-75k characters. So I can do roughly 15 HOURS per month with a full stack. That's a great deal.

My fears?

1) They voices never improve? Many are trash, but some are solid, you DO have to play with them a bit (They let you preview for free)

2) The software dies (like Speechki, which was the BEST and which I'm dying to replace; or Uttr which became Blakify, and then subsequently died). 1.1M characters per month - if we use it - isn't free. They're banking on this base helping them launch. I hope they're successful (that's why I'm leaving them a positive review). But Appsumo's options currently are batting only 3/5 (and Play.ht seems to have depricated some things for sumos).

I've also purchased a full stack of VoiSpark, which has higher quality voices, but currently has a STAGGERINGLY SMALL 10k character limit, which is like 6-8 minutes. That migght be great for some poeple, but not me (I'm TOLD they're raising that, possibly this month (Sept. 2025) - but that is yet to be seen.).

So...

I'm hopeful. I really do want to stack more codes or - if I have to and am allowed - buy a second (or third) full stack. This combined with a platform like Subscribr (also found via Appsumo, and probably in my top 10 all time deals) is KILLER for making a podcast that has legs. If VoiSpark fixes the low character limit, I might 2x or 3x that deal, because they're that useful and will be for the foreseeable (but probably not indefinite) future.

You can get this deal in good conscience.

(So I wrote this review, but accidentally put it on VoiSpark - they're hard to keep straight!)

No background music - only sounds available for Text to speech
App does ok, but there is still a lot of room to get better. Almost every other text to speech bot (also free ones) offers a big variety of different background music. Why do you only have sounds and no music?

I purchased Plan 2 of Unmixer on AppSumo because the description clearly states it includes the feature “Translation and text transformer, Import Documents in Voice Studio,” meaning the ability to translate and import full documents. My intended use was exactly that: to translate documents and listen to them in a human-like voice for study.

To my surprise, when I entered the application I found that this feature is locked. To activate it, I am required to purchase an additional $5/month subscription, which contradicts the AppSumo offer.

This is false advertising and a hidden upselling practice: it is advertised as included in Plan 2, but then an additional monthly payment is required.

I feel misled, because I bought this plan to solve a specific need (document translation), and that feature was the main reason for my purchase. I recommend other buyers review carefully before deciding, and I hope AppSumo takes action to ensure Unmixer delivers what was promised.

👉 I need a solution to this issue and a clear response from the Unmixer team.

Pros:
• For Tier 2 and above, access of more high-quality models (‘multilingual’ + ‘express’ models)
• Long Form Studio impressively supports ~200K char per generation (although like other AI providers including OpenAI and Artlist, it also occasionally “skips” some sentences and “repeats” some sentences).
• Intuitive UI for Dubbing Studio, making editing translation easier
• Good voice-cloning quality
• BYOK OpenAI for consistent character if “full-range-emotions” is needed
• Some additional perks like copywriting templates, audio transcription
• Good support & good-will of Founder

Cons:
• API TTS is limited to 2K char per generation (vs 5K from OpenAI; ~10K from ElevenLabs). (But this has been promised to increase over time)
• Occasional API instability (but it often resolves on its own after a time)
• Dubbing has inconsistent fluctuating paces (although they’re working on improving this)

Verdict: If you don’t need that highly advanced full-range emotions for film-making (which ElevenLabs would be more suitable), this is a GOOD INVESTMENT at LTD price and AFFORDABLE at monthly subscription prices.