Synthesia's video editor needs to be improved. We're working with our Synthesia rep to get this issue at the forefront of the developer's minds. I haven't used it in a few months, but the last time I picked it up, the video editor wasn't that great. Synthesia should include a good audio editor and a better timeline UI.
I would like to have a little bit more of a place to have testing be able to test more because we're being charged on a utilization level. And sometimes, we have to scrap what we created. It'd be nice to just use that quota for actual production videos and have something. It would be nice to just have an audio track to check the audio because we're in a technical field, and the avatars' voice rendering gets strange on certain words that we use. It gets strange with certain words that we use. We use a lot of acronyms and it's just not pure English. So it'd be nice if we didn't lose our credits on testing rather than if we just had an audio tracker to cut a lot of that stuff. So, the development testing workflows could have been improved at that time. With the voice rendering, if we could have some type of AI that could understand the state, and the company, it could probably make a lot of improvements on the language, for clarity, for spelling, it could check a lot of that stuff for us. If it had some type of intelligence to it, we'd be able to pronounce certain things and understand the context of certain things. So, just a level of smarts in there would be a huge improvement.
Some of the AI models can be improved, but not all of them. Some look less human-like, especially the older models. The newer models are great. Synthesia's compatibility with different learning patterns could be improved. I use it with Squarespace, and it's easy, but it could be better with other elements.
One of the issues we face is the variability in accents; there are hundreds to choose from, but some work better than others, and pronunciation can sometimes be subpar. Facial expressions may also not always be entirely accurate. Providing the capability to feature two avatars on-screen simultaneously for interactive communication would be a valuable enhancement. Improved transitions between different segments of the content would also enhance the overall experience.
One aspect I'd like to see improvements in is the customization of avatars. Often, clients want to use their own faces or company representatives as avatars, but they find the current pricing of a thousand dollars for a custom avatar to be too expensive. This cost is a turn-off for them. I hope you can work on this by either reducing the cost or finding a solution to make it more affordable for users who wish to have a personalized avatar. One concern is related to the cost, which some clients have raised. The second issue pertains to moderation and time zones. I believe they should address this concern because, recently, during testing, I encountered a situation where customer service was unavailable to manually review a video demo. This manual review process is important to ensure that the video complies with their terms and conditions and doesn't violate any guidelines. The video review process can take around five minutes, and it's crucial to ensure that users are creating content within the specified terms and conditions.
Choose from 150+ ethnically diverse stock AI Avatars.
Create videos in 120+ languages, accents and voice tones.
Simply type in your text and turn it into professional voiceovers in minutes.
Create your own custom AI Avatar (your digital twin).
Add micro gestures to your avatar such as head nods or raised eyebrows.
You can clone your own voice and pair it with your own custom AI Avatar.
Synthesia's video editor needs to be improved. We're working with our Synthesia rep to get this issue at the forefront of the developer's minds. I haven't used it in a few months, but the last time I picked it up, the video editor wasn't that great. Synthesia should include a good audio editor and a better timeline UI.
I would like to have a little bit more of a place to have testing be able to test more because we're being charged on a utilization level. And sometimes, we have to scrap what we created. It'd be nice to just use that quota for actual production videos and have something. It would be nice to just have an audio track to check the audio because we're in a technical field, and the avatars' voice rendering gets strange on certain words that we use. It gets strange with certain words that we use. We use a lot of acronyms and it's just not pure English. So it'd be nice if we didn't lose our credits on testing rather than if we just had an audio tracker to cut a lot of that stuff. So, the development testing workflows could have been improved at that time. With the voice rendering, if we could have some type of AI that could understand the state, and the company, it could probably make a lot of improvements on the language, for clarity, for spelling, it could check a lot of that stuff for us. If it had some type of intelligence to it, we'd be able to pronounce certain things and understand the context of certain things. So, just a level of smarts in there would be a huge improvement.
Some of the AI models can be improved, but not all of them. Some look less human-like, especially the older models. The newer models are great. Synthesia's compatibility with different learning patterns could be improved. I use it with Squarespace, and it's easy, but it could be better with other elements.
One of the issues we face is the variability in accents; there are hundreds to choose from, but some work better than others, and pronunciation can sometimes be subpar. Facial expressions may also not always be entirely accurate. Providing the capability to feature two avatars on-screen simultaneously for interactive communication would be a valuable enhancement. Improved transitions between different segments of the content would also enhance the overall experience.
They should indulge in better advertising and also voice modulation features for the videos should be introduced.
One aspect I'd like to see improvements in is the customization of avatars. Often, clients want to use their own faces or company representatives as avatars, but they find the current pricing of a thousand dollars for a custom avatar to be too expensive. This cost is a turn-off for them. I hope you can work on this by either reducing the cost or finding a solution to make it more affordable for users who wish to have a personalized avatar. One concern is related to the cost, which some clients have raised. The second issue pertains to moderation and time zones. I believe they should address this concern because, recently, during testing, I encountered a situation where customer service was unavailable to manually review a video demo. This manual review process is important to ensure that the video complies with their terms and conditions and doesn't violate any guidelines. The video review process can take around five minutes, and it's crucial to ensure that users are creating content within the specified terms and conditions.
The video part needs improvement.