from what I can see hedra has a hard time with smiling faces, as long as a face is serious, it doesn't matter whether it's a real person or a painting, it'll do a great job.
Guys, that is AI guys, you must stimulate AI consumption. Put together a prize, a convention or whatever else, create incentive for serious AI creation. (substantial, goal oriented, competition oriented).
if loading your own audio or music you are limited to 30 seconds audio. but it is great but i cant wait for the full version.. with improved graphics and upscaling
Hedra is great for talking not so much for singing, unless you want to pay for enough time to do at least a 3 minute song, or unless you want to cut your 3 minute song into 30 second chunks then put them back together
I really like the lip synch timing of Hedra. The terms of use, however, indicates that any uploaded image must be mine, and yet, this RUclips video has other AI images that are talking avatars. Am I misinterpreting the TOU. Has the TOS been updated to allow AI created images to be uploaded?
Microsoft announced that VASA-1 is only a research demonstration and there's no official product or API release plan yet. This may change in the future though so we'll have to wait to use it.
Yes, you can vary the voice options by clicking the "Import Audio" tab and uploading your own audio file (it has to be MP3 or wav file). You can create your own audio file by using an AI voice generator like Murf, Canva or Elevenlabs (I have link in the description). OR you can just record your own voice and upload that to Hedra. I agree though, the default voice options are limited (and do kinda sound like American TV lol). Hopefully, they add more voice options to the "Generate Audio" feature through Hedra.
Always messes up your input character image, can't make a clean version of anything, I don't want my frigging artwork to look like a stupid potato or rabbit☹️
It's not a God tier until we can put our own sheet music to it. A God tier would be able to put to Python and create the voice from the pc's own generated tone. God tier won't need voice actors.
“I created that myself”, after clicking on a generate button using AI, will never come across my mind to say as an artist, ever… I will simply state “I created that using AI”. Feels a bit more honest/genuine…
I have been in touch with Hedra regarding this. They informed me that they're working to ensure the app flow complies with Texas biometric privacy laws. They’re working on getting it available as soon as possible and there will be more updates within the week.
Sally - very realistic, superb. Female with American accent. Relaxed voice with good conversations. elevenlabs.io/app/voice-lab/share/899444b1d9eab0cea6d20686befafde9ccdc46ab4628e79973cf318a9497b138/KFa4YZbTDlxigRroAoMz
I have been in touch with Hedra regarding this. They informed me that they're working to ensure the app flow complies with Texas biometric privacy laws. They’re working on getting it available as soon as possible and there will be more updates within the week.
from what I can see hedra has a hard time with smiling faces, as long as a face is serious, it doesn't matter whether it's a real person or a painting, it'll do a great job.
Guys, that is AI guys, you must stimulate AI consumption. Put together a prize, a convention or whatever else, create incentive for serious AI creation. (substantial, goal oriented, competition oriented).
if loading your own audio or music you are limited to 30 seconds audio.
but it is great but i cant wait for the full version.. with improved graphics and upscaling
Same here, I can't wait to see what it can do in the future.
Hedra is great for talking not so much for singing, unless you want to pay for enough time to do at least a 3 minute song, or unless you want to cut your 3 minute song into 30 second chunks then put them back together
It has censors. You cant just say anything. But it is the biggest breakthrough in A.I since the arrival of Chatgpt.
This is such a useful tool to add to ones arsenal. Thanks for sharing.
I really like the lip synch timing of Hedra. The terms of use, however, indicates that any uploaded image must be mine, and yet, this RUclips video has other AI images that are talking avatars. Am I misinterpreting the TOU. Has the TOS been updated to allow AI created images to be uploaded?
Ilona Muskia in action ^^
Looks like a useful tool to have. I will give this a try. Thanks 4 sharing!
Hello Tried 9 times and each time video generation failed. typed in the prompt for speech generate an Ai pic and then selected generate and failed.
G Tier, any idea when Hedra will fix the blurry output after generating the video?
Imagine when you can tie it to things like grok at super speed to create interactive avatars
Great video,Thanks for posting.
Appreciate it, thank you for watching.
I want download it
How do you make it sing!!??
At 12:07, am I the only one thinking woman Elon Musk looks like Sigourney Weaver?!!!
the link provided for vasa-1 shows description not the actual app. where can I access Vasa -1?
Microsoft announced that VASA-1 is only a research demonstration and there's no official product or API release plan yet. This may change in the future though so we'll have to wait to use it.
Great video! Cant wait to try this!
Music video created with Hedra:
ruclips.net/video/2W7_hHWPoxA/видео.htmlsi=kUrT6vp2Tnt-9OtC
Hedra is it possible to vary the voice options. We can't have every every voice sounding like it came from American TV
Yes, you can vary the voice options by clicking the "Import Audio" tab and uploading your own audio file (it has to be MP3 or wav file).
You can create your own audio file by using an AI voice generator like Murf, Canva or Elevenlabs (I have link in the description). OR you can just record your own voice and upload that to Hedra.
I agree though, the default voice options are limited (and do kinda sound like American TV lol). Hopefully, they add more voice options to the "Generate Audio" feature through Hedra.
Always messes up your input character image, can't make a clean version of anything, I don't want my frigging artwork to look like a stupid potato or rabbit☹️
It's not a God tier until we can put our own sheet music to it. A God tier would be able to put to Python and create the voice from the pc's own generated tone. God tier won't need voice actors.
Waiting for something better to hit open source, instead of funneling money to a platform that's already started down the road of enshittification.
Seriously can't wait for an open source version of this!
Same, and I think open source will come... *eventually.*
You mean paid/closed open source.
such a cool tool
Useful video. Thanks
Too much blinking is messing it uo
“I created that myself”, after clicking on a generate button using AI, will never come across my mind to say as an artist, ever… I will simply state “I created that using AI”. Feels a bit more honest/genuine…
only 1 minutes allowed in free version ..
Doesn’t work for free 😅😅😅
This is the best AI introduction I've seen so far. Keep it up!
Thanks, will do!
Thank you.
So is it a porn site? Why is it's use blocked in Texas and other states?
Wondering same thing.
I have been in touch with Hedra regarding this.
They informed me that they're working to ensure the app flow complies with Texas biometric privacy laws.
They’re working on getting it available as soon as possible and there will be more updates within the week.
tried it. it sucks.
It's hit or miss for me. It has potential though.
It's going to cost big time. Its good.
I agree.
when do we get a nsfw version? otherwise it's useless.
what voice was used to voicoiver this video?
Sally - very realistic, superb.
Female with American accent. Relaxed voice with good conversations.
elevenlabs.io/app/voice-lab/share/899444b1d9eab0cea6d20686befafde9ccdc46ab4628e79973cf318a9497b138/KFa4YZbTDlxigRroAoMz
Why are some US states restricted? Texas residents? Why?
I agree! I live in the great state of TX, so I guess I'm restricted! Oh man!
I have been in touch with Hedra regarding this.
They informed me that they're working to ensure the app flow complies with Texas biometric privacy laws.
They’re working on getting it available as soon as possible and there will be more updates within the week.