HeyGen Reviews & Product Details

What I like most about HeyGen is that I could import a voice I created in another program, so I didn’t have to recreate it from scratch. I also really appreciate how realistic her voice sounds, especially with the enhanced emotion, and how clearly her mouth articulates the words when she reads the scripts I write. Review collected by and hosted on G2.com.

What I don’t like is that there isn’t a way to control who’s speaking in a clip when more than one character is visible. After several attempts (and wasted credits), I still couldn’t get the results I wanted and ended up settling for a static image with an audio track instead. That kind of soured my Mother’s Day clip. Review collected by and hosted on G2.com.

I love the AI agent in HeyGen. It helps me avoid burnout and fatigue because I don't have to repeat myself constantly. The ability to edit scene by scene is great, and I don't have to start over entirely. The ease with which I can reduce 50 hours of training to just 5 hours of work on a video is impressive. The initial setup was very easy, and I was able to integrate it with ChatGPT smoothly. Review collected by and hosted on G2.com.

I find it takes me a few tries to get overlays or pronunciation right. I would keep them more relevant to the training. Review collected by and hosted on G2.com.

I love the avatar quality and realism in HeyGen, especially the custom instant avatars that are shockingly lifelike. The voice cloning is impressive as it captures my actual tone, pacing, and emotion. The no-shoot workflow is fantastic; I can go from script to avatar video in minutes without ever touching a camera. Also, the initial setup was very easy, making it really accessible. Review collected by and hosted on G2.com.

While HeyGen is a powerful tool, there are a few areas where users, myself included, see room for improvement. These often come down to the realism of the avatars. Although the newest avatar V model is a significant leap forward in realism, further improvements are needed. The areas of improvement are not about fundamental flaws but rather fine-tuning the nuances of human expression and performance. While avatar V can generate micro-expressions, users have noted that avatars can feel stiff, emotionless, or lack the warmth of a real person. Review collected by and hosted on G2.com.

I used HeyGen in a course on AI held at the University of the Third Age (UNITRE) in Turin (Italy). Particularly useful was the ability to use different Avatars (myself, AI-generated figures) assigning them brief spoken interventions with an impressive presence effect and lip synchronization. I also produced short films where I mixed my Avatar with clips extracted from science fiction movies useful for explaining AI behaviors. Review collected by and hosted on G2.com.

I had some difficulty in obtaining a video that adhered to the related script: HeyGen showed a certain tendency to interpret the instructions in its own way. However, I believe this is part of the normal learning curve. Review collected by and hosted on G2.com.

What I like most about HeyGen is how quickly it allows creative ideas to become cinematic experiences. The new Workbench concept is especially powerful because it helps organize scene structure, emotional pacing, and visual consistency in a much more natural way. It feels less like using separate AI tools and more like directing a real creative production workflow. Review collected by and hosted on G2.com.

Sometimes generating high-quality cinematic results can take multiple iterations and prompt refinements, especially for more complex storytelling scenes. However, this is currently common across most AI video generation tools. I would especially love to see faster rendering times, improved scene consistency between shots, and even stronger control over cinematic continuity and object persistence in longer creative workflows. Review collected by and hosted on G2.com.

Ease of use, ease of implementation, customer support, frequency of use, number of features, and ease of integration. Review collected by and hosted on G2.com.

The biggest issue is transparency around feature limitations and beta status.

For example:

* Seedance Avatar Shots does not support Voice IDs or voice clones, but this is not clearly communicated before generation.

* Presenter Avatar V appears to still be in beta, but there is no obvious beta label or warning visible during usage.

* Some experimental features can produce highly unpredictable outputs, including incorrect duration or unexpected speech behavior, while still consuming large amounts of credits.

For users working on professional projects, especially cinematic or timing-sensitive videos, these limitations matter a lot because they directly affect planning, workflow, and credit usage.

I also think there should be clearer expectations around:

* which features are stable for production use

* which features are experimental

* what kinds of outputs are currently unreliable

* whether timing controls are approximate or strict

Another issue is that some features feel visually connected inside the platform, which creates the expectation that workflows are compatible across tools, even when they are not. For example, using avatars + voice clone + cinematic generation feels like a natural expectation from a user perspective. Review collected by and hosted on G2.com.

I like HeyGen's easy app access, which helps me understand how to get things done efficiently and in a quality way. I also appreciate that I don't have to put in a considerable amount of time for each video because HeyGen helps to create those aspects smoothly. The initial setup was very simple, and it wasn't hard at all. Review collected by and hosted on G2.com.

Sometimes when there's a new product release, it takes a while to work out the bugs. It feels like they're still charging us as if it's a full version when it seems more like a beta version. The credits should not be deducted when testing these out before everything is in order. Review collected by and hosted on G2.com.

I like that HeyGen is easy to use for creating professional-looking academic tutorial videos without needing a full studio setup. I use it for educational content, including Research Methodology, Biostatistics, manuscript writing, faculty training, and student learning support in allied healthcare sciences.

The avatar and AI video workflow is helpful because it lets me turn prepared teaching scripts into clear video lessons. It is especially useful when I need to explain repeated academic topics in a consistent and engaging way for students and faculty. For educators, this saves time and makes it easier to prepare structured tutorial videos, orientation content, and training material.

The right-side “Overall Review Depth” is only showing which topics your answer covers. You don’t need to separately answer those points now. Review collected by and hosted on G2.com.

The main area I would like to see improved is clearer information on credit usage, especially when creating longer academic videos or using PPT/PDF-to-video workflows. Sometimes it is difficult to estimate how many credits a project will need before generating the video.

For educators, more ready-made templates for teaching, tutorials, orientation sessions, and academic training would also be helpful. Long lesson workflows could be smoother, especially when working with multiple scripts, slides, voice files, and avatar videos. Overall, these are not major issues, but improving them would make HeyGen even more useful for educational users. Review collected by and hosted on G2.com.

The most important thing for me is that the number of videos is unlimited for a subscription of around 20 dollars. And it's very important, if it wasn't unlimited, I would never have subscribed. Review collected by and hosted on G2.com.

What I don't like is that we can't download in wav or mp3 a text that we just put into a template. Because often there are many pronunciation errors regardless of the voice model we use, and we have to manually redo it by recording patches with our own voice for editing. If we could download the sound, it would avoid a lot of CPU and GPU overload. Because for now, I am forced to generate a video and then extract the sound from it to modify it. An export in wav or MP3 would be wonderful. Review collected by and hosted on G2.com.

What stands out most about HeyGen is its uncanny ability to handle lip-syncing and voice cloning with impressively high fidelity. For anyone focused on narrative structure and technical efficiency, it’s genuinely fascinating to watch how the AI translates gestures and micro-expressions so fluidly and convincingly. To me, it feels like a game-changer for scaling high-quality audiovisual content, helping ensure that a creative vision isn’t held back by the logistical constraints of traditional production. Review collected by and hosted on G2.com.

While the technology is genuinely impressive, the pricing structure can feel quite restrictive—especially when you’re working on extensive narrative projects that demand multiple iterations. The credit system, in particular, seems to run out too quickly, which becomes frustrating when you’re trying to fine-tune micro-expressions or lock in a very specific delivery and tone. And although the AI is clearly advanced, there’s still a slight “uncanny valley” quality in certain movements that can pull you out of the moment and make the result feel less organic. In those cases, it often takes extra post-production work to ensure the final output fully matches a high-standard creative vision. Review collected by and hosted on G2.com.