HI there! Is there something that can create a talking avatar from my pets photo? I have not found anything! Maybe you know of something that will work for dogs?
Thank you for the great tutorial! How did you get Dall-E to generate the same character more than once? Every time it generated for me a different looking character.
just tested it on many examples and because of these small inconsistences between images variations (in many cases actually big inconsistencies) it is not possible to use generated images anywhere without lots of photoshop work - I think we are still not there with AI generation for this type of work
It depends on what you are working on. If you are working on images for a children's book the small inconsistencies are not going to make a difference at all. Very few children would even realize there are inconsistencies. Some of you over think this stuff. If done right this procedure can be extremely consistent.
I totally agree, my three tries were total fails. Even when asked to change the image, all it did was tilted the image to a 45 degrees. I think i might try to make my own gpt!
Zinny always giving us value! pls guys ensure you watch the ads on the video... its the least support we can give her... thanks Zinny, learnt quite a bit from your channel
Thank you for creating this video. I am a writer and I do draw, but I wanted to dabble with AI and it was not as good as I could draw (for animation) I was frustrated with my outcomes from all the cool tools. But this video not only helped me make consistent characters, but now I can move forward with my project and tell my tales. I am also going to start making comic books for teens and even a CYOA project is now a go. If you ever decide to sell courses or do live trainings, I would refer folks. Thank you so much.
you just picked up a new subscriber-the information you shared is greatly appreciated for I been struggling a bit with consistent characters for my children's book. Your instructions were easy to follow- thanks for the video
That animated character was the coolest thing Ive seen in awhile. Did you create that using chatgpt? Can you make a tutorial on how to animate characters to talk?
Thank you. No i didn't use Chatgpt i used Midjourney. I am working on a tutorial to showcase how i created the Avatar and how to make it talk. So stay tuned for that
That was a good presentation thank you! I subscribed. By the way, do you have any video already posted showing how to have an animated Pixal character just like yours in this video? It's awesome and I think that the lips movements are really, really good.
This is awesome thank you so much for the video! I've been waiting for this for so long ahhaa btw when you wanted a picture of him "Walking" you were pronouncing it and typing it "working". Not trying to be mean just trying to help a little = ] Keep up the good work!!!
Mons!!! 👏👏👏 In the last few months I have learned I have a super talent...I seem to remember the speech mannerisms and cadence of every person I meet not only in real life, but also over the interwebz.
Thank you for the tutorial. I wished I had come across this months ago, before I created my coloring book. But, I still have ideas for future projects.
This is really great, thanks so much for sharing! I’m thinking of creating an ASMR channel or something like that but I would like to animate it and just add the sounds
Great content, only question is how to converted to compressed was it? when you downloaded if from ChatGBT then you compressed? to keep the quality BEFORE you upload it to Canva. A little lost there... Thanks!
You can use bigjpg.com. It is very simple, when you get to be website, upload the image you want, then chose how many times you will like to upscale the image. When it is done you will download it and use
I definitely have a question. I am new to creating children's books. I did some of the stuff you said, took the background out took the image, and put it in something else but when you upload it on Amazon, it takes the image out of the picture. how do you stop that cause that is the problem that I'm having?
Didn"t Chat gpt just hook up with Canva?? So their subs could just use it for prompt writing?? It is costly to pay for 3 different generators that do the same thing. Dall -E is owned by Chat. So why should subscribers pick both ??
So, there are bunch of GPTs that hook up to existing tools. But for the sole purpose of serving as a quick link to the other tools and not a replacement. For you to even link properly you need to have account in both tools. If you want to do something else serious for example with Canva, you need to have a canva account. If you want to use a feature in canva that is paid. You will need to pay for canva pro. Having a Chatgpt subscription doesn't automatically cover the paid version of other tools. Hope this is clear enough
Awesome. Thank you, Zinny! Do you know a way, after I’ve created a figure and some days have passed, that I can go back and attach the image to have a different pose? When I try to do it (I'm not in the original GPT chat), it changes the figure.
Thank you for this great video, I have a question though. How did you downloaded the pictures in png format and how did you upscale them before that? If I right click on the images in chatgpt, I can only download them as.webp files.
Great ... where have you been all my life? Seriously, ... you get into exactly what we need to take on the established Movie Industry (Hollywood as it were etc.). I don't know if you realize it or not, but these techniques are a lot better than picking cotton and sugar cane. I love what you have revealed. I remember pencil and paper (if that was available) LOL Yes Thank You Zinny
Can you please show how you create the opening of your video? I mean, how do you create the animated girl to talk like that? How do you take an image and make it into a video with lip sink audio? Thank you
Thank you, your explanation is very detailed and easy to understand. I want to ask, what is the command to GPT to draw two people at once in one frame, for example Sam riding a bicycle carrying his younger brother Andi?
I haven't tried multiple characters in a scene yet. I will look into it and see if is possible at the moment and if it is i will definitely make a tutorial about it
Thanks for the great video. I just signed up after watching your video, however, it was a total fail for me. I’ll post my video soon and link to you when I’ve uploaded my results.
Very informative video. Thanks. I have a question. 1) You created "Sam" and made a couple of images. 2) Then, you created a new charactor named "Olivia" and made some images. 3) If you want to make more images of "Sam" the next day, will they keep the consistency as the day before?
Thank you for your effort and tutorial, how do you keep character consistency like, if we open a new chat on GPT (let’s say the old chat was deleted), how do you continue to generate same character again in new chat with something like, GEN ID to recall the character? Thank you for your time and looking forward to see you here.😊
I followed along and when I told it to design a 2D animation style, after a couple attempts, it gave me something perfect. I then told it so and that I wanted to see him in a different pose. The next result was a completely different design, like...dramatically different. I tried referencing the previous image, and it was still completely different again. I tried running it through an image describer and try and guide Consistent Character back to where I needed it but no matter what, it never generates anything the same.
This is a wonderful video, thanks for sharing. I learned many useful insights. Will have to sign for GPT Plus, I have been delaying the decision, but now... Great!
Hi Zinny, In this video at timestamp 10 mins : 56 seconds, You have upscaled the images. Could you please specify the site you have used to upscale and is it free for multiple images ? Waiting for your response. Thanks.
I dont have a separate tutorial for upscaling. But is quite easy. Just visit bigjpeg.com. Select the images you want to upscale, chose how times you want to upscale, then upscale, download and use.
Something is wonky. When I try to download the images of my consistent character, they're downloading either in .xml or .webp Any clue? I tried asking the GPT and could never get it to download in an image format. I'm on my Lenovo Ideapad. Have never experienced a problem like this before.
Actually is not a problem. Chatgpt change the image format long time ago to webp. All you need to do is use a convert to convert to either jpeg or png.
@@thezinnystudio I finally found some place that talked about Google deciding they wanted smaller file sizes. I just pull them into Photoshop and change them now. Interesting though, I was working on a consistent character this afternoon. ChatGPT misunderstood my directions and created a landscape image with the character in different poses all on the same illio. I asked it to separate them and show them vertically, it did, and created them as PNGs. I had not expected that.
Hmm, interesting. I didn't do any training before hand. As you can see in the tutorial, the images were generated live. Try this to see if it works, back track to the last prompt before the no consistent image was created, edit the prompt and specify that image should be created with the last image generated without modification. That way chatgpt will know you are referring to the very last image you created and want to continue using.
Yes and No. Yes in the sense that it will create an image using the picture as a reference. but it wont be exact replica that's the no part. But you can give it a try
@@thezinnystudio . Ah yes. Sorry, as I'd watched the video and didn't hear it there and was just a bit surprised to see the very similar character prompts. No worries. Best of luck with the channel.
It's a great informative video, is there any scope to create same consistent characters of entire family persons like 4 to 6 members in a family? for a story telling video
thankyou for this . Greast Video. When i use this GPT it will only let me fdo 4 variations of the character, Is there any way to prevent this restriction?
@thezinnystudio Great tutorial! Well explained and demonstrated. I am having trouble when requesting another expression. Consistent Characters GPT keeps giving me different characters and even races although I am specifically asking to use a particular image that it created. Please advise, and thank you in advance for your consideration regarding this matter
Not working for multi characters : I'm here to focus on character designs and variations based on specific prompts about individual characters. If you're looking to illustrate scenes involving environments, interactions between characters, and detailed settings, I might not be able to fulfill this request directly. However, I can help create images based on descriptions of characters or their actions within a scene. Would you like to continue with character design or explore a different aspect of your story's visualization?
I hope you can help, I watched your video and then purchased a Chat GPT pro plan for $20 for a month. Now I am unable to access any Open API apps or get to the screen you are showing. Is this something about buying credits?
OK. Just wait abit, I am not sure if for new account if it takes time to show up. But I will research abit on it. You shouldn't have to do anything else.
![Noob To Max With DRAGON REWORK In Blox Fruits [FULL MOVIE]](http://i.ytimg.com/vi/LnBMOinoOvA/mqdefault.jpg)
Thank you for watching. Have a questions leave it below👇.
great video ! which tool did you use to upscale your images
there is a free upscaling tool called Upscayl which I use on both mac and windows and it is not web based.@@oneroad2930
HI there! Is there something that can create a talking avatar from my pets photo? I have not found anything! Maybe you know of something that will work for dogs?
Thank you for the great tutorial! How did you get Dall-E to generate the same character more than once? Every time it generated for me a different looking character.
just tested it on many examples and because of these small inconsistences between images variations (in many cases actually big inconsistencies) it is not possible to use generated images anywhere without lots of photoshop work - I think we are still not there with AI generation for this type of work
Come up with yours then😂😂😂 come on man. Learn to appreciate.
It depends on what you are working on. If you are working on images for a children's book the small inconsistencies are not going to make a difference at all. Very few children would even realize there are inconsistencies. Some of you over think this stuff. If done right this procedure can be extremely consistent.
I totally agree, my three tries were total fails. Even when asked to change the image, all it did was tilted the image to a 45 degrees. I think i might try to make my own gpt!
@@jamesdavis8731if you want to create real wealth then quality matters.
I don’t see any inconsistencies with the character in the video
Zinny always giving us value! pls guys ensure you watch the ads on the video... its the least support we can give her... thanks Zinny, learnt quite a bit from your channel
Thank you
You just made me rich through this videos, great content ! thank you !
Thanks
Thank you so much
Thank you for creating this video. I am a writer and I do draw, but I wanted to dabble with AI and it was not as good as I could draw (for animation) I was frustrated with my outcomes from all the cool tools. But this video not only helped me make consistent characters, but now I can move forward with my project and tell my tales. I am also going to start making comic books for teens and even a CYOA project is now a go. If you ever decide to sell courses or do live trainings, I would refer folks. Thank you so much.
you just picked up a new subscriber-the information you shared is greatly appreciated for I been struggling a bit with consistent characters for my children's book. Your instructions were easy to follow- thanks for the video
You are welcome
I just went along with it. It was really fun and interesting. Thank you!
Glad you liked it
Thanks Zinny for the tutorial. Have been waiting for it
Amazing #now I can design new storybooks for my kids. 😊
Thanks for your efforts, subscribed!
Uwc
Hi Zinny, your content is great! I would just suggest to put transcriptions on each video. It is good if someone is not able to listen to the audio.
This is what I needed. Thank you so much value!
That animated character was the coolest thing Ive seen in awhile. Did you create that using chatgpt? Can you make a tutorial on how to animate characters to talk?
Thank you. No i didn't use Chatgpt i used Midjourney. I am working on a tutorial to showcase how i created the Avatar and how to make it talk. So stay tuned for that
OMGGGGGG I LOVE THIS, GOOD INFO WORKING ON MY CHILDREN’S BOOK 📚
Glad it was helpful!
That was a good presentation thank you! I subscribed. By the way, do you have any video already posted showing how to have an animated Pixal character just like yours in this video? It's awesome and I think that the lips movements are really, really good.
I have a video where I showed how to create a talking Avatar. You can check it out
@@thezinnystudio I will. Thanks
This is probably one of the BEST videos I’ve watched on RUclips in quite some time
Thanks
I agree
Awesome content you have created. I have just subscribed to your lovely channel. One question, all these images will be published in jpeg ? Or?
Yes
You are amazing! Thank you for sharing.
Thanks for watching!
This is awesome thank you so much for the video! I've been waiting for this for so long ahhaa btw when you wanted a picture of him "Walking" you were pronouncing it and typing it "working". Not trying to be mean just trying to help a little = ] Keep up the good work!!!
Very thorough! Thanks 🎉
You are welcome
Thank you so much for sharing. very helpful
Glad it was helpful!
Fantastic video! Thank you so much for sharing these instructions!
Uwc
Very good. I learned something new!!! Thank you.
You are welcome!
Very good video, a very simple way of teaching with Hands-On-Practice
Thank you
Thank you so much!!!
Uwc
Mons!!! 👏👏👏 In the last few months I have learned I have a super talent...I seem to remember the speech mannerisms and cadence of every person I meet not only in real life, but also over the interwebz.
Love the video, thanks heaps for it. What do you use to upscale the images?
bigjpg.com
Thank you for the tutorial. I wished I had come across this months ago, before I created my coloring book. But, I still have ideas for future projects.
Uwc
This is really great, thanks so much for sharing! I’m thinking of creating an ASMR channel or something like that but I would like to animate it and just add the sounds
That's good
Thank you so much such a great video, I had to subscribe. May you please do a video on how the characters are talking? Thank you so much again.
I have a video already on that, check it out on the channel.
Excellent,Maybe video showing how to put an ebook together,tried couple but you explained so well
Noted
Thank you this has been one of the best I have seen great job
Thank you
This is brilliant, and it's giving me ideas.,Thank you
Uwc
How do I reference the illustrations in publishing the book?
Please paraphrase, didn't get the question
@@thezinnystudio do I need to add that DALLE-3 is my illustrator in publishing the book?
Or is there no need to write a illustrators when it’s AI generated?
Great content, only question is how to converted to compressed was it? when you downloaded if from ChatGBT then you compressed? to keep the quality BEFORE you upload it to Canva. A little lost there... Thanks!
when you download the image. you would have to upscale it, like every other Ai generated images. Then you can go ahead to use in canva
@@thezinnystudio how do you upscale it ? TIA
You can use bigjpg.com. It is very simple, when you get to be website, upload the image you want, then chose how many times you will like to upscale the image. When it is done you will download it and use
Does this GPT work with images we upload? Like, if we wanted to use an imagine we already created, or do I need to prompt it?
I tried it but, it doesn't give the exact as the picture but uses the features to create similar image. You can give it a try and see for yourself.
I definitely have a question. I am new to creating children's books. I did some of the stuff you said, took the background out took the image, and put it in something else but when you upload it on Amazon, it takes the image out of the picture. how do you stop that cause that is the problem that I'm having?
Didn"t Chat gpt just hook up with Canva?? So their subs could just use it for prompt writing?? It is costly to pay for 3 different generators that do the same thing. Dall -E is owned by Chat. So why should subscribers pick both ??
I’ve been wondering the same thing
So, there are bunch of GPTs that hook up to existing tools. But for the sole purpose of serving as a quick link to the other tools and not a replacement. For you to even link properly you need to have account in both tools. If you want to do something else serious for example with Canva, you need to have a canva account. If you want to use a feature in canva that is paid. You will need to pay for canva pro. Having a Chatgpt subscription doesn't automatically cover the paid version of other tools. Hope this is clear enough
Can you do a video how you created your character and make the lips match what you're saying. I think it's so cool
Tutorial coming soon
Awesome. Thank you, Zinny! Do you know a way, after I’ve created a figure and some days have passed, that I can go back and attach the image to have a different pose? When I try to do it (I'm not in the original GPT chat), it changes the figure.
You will have to work within the original message, so you can refer back to pervious images.
@@thezinnystudio Thank you for your response.
Thank you so much ❤ it's helpful 🎉
You're welcome 😊
Omg, this is exactly what I needed! TFS
Glad the tutorial helped
I'm not sure about the topic. I am planning to purchase chatGPT-4, however whose copyright does the images I create using the GPT-4 tool belong to?
Thank you for this great video, I have a question though. How did you downloaded the pictures in png format and how did you upscale them before that? If I right click on the images in chatgpt, I can only download them as.webp files.
OMG!!!!! U gave us such valuable info---thank you sooooooooooooo much
Great ... where have you been all my life? Seriously, ... you get into exactly what we need to take on the established Movie Industry (Hollywood as it were etc.). I don't know if you realize it or not, but these techniques are a lot better than picking cotton and sugar cane. I love what you have revealed. I remember pencil and paper (if that was available) LOL Yes Thank You Zinny
UWC
Awesome video. Thank you.
Uwc
Can you please show how you create the opening of your video?
I mean, how do you create the animated girl to talk like that?
How do you take an image and make it into a video with lip sink audio?
Thank you
There is a video on the channel called creating a talking avatar.
thanks and good video
so since you gpt4 do they allows you to apps in the gpt store with subscription for their full versions
nice work bro i like it😍
This video was very informative. Thank you so much! Can you please clarify what program you used to upscale the images?
love the video. what are you using to upscale your designs?
I use bigjpg.com
Thank you, your explanation is very detailed and easy to understand. I want to ask, what is the command to GPT to draw two people at once in one frame, for example Sam riding a bicycle carrying his younger brother Andi?
I haven't tried multiple characters in a scene yet. I will look into it and see if is possible at the moment and if it is i will definitely make a tutorial about it
I look forward to it
Thanks for the great video. I just signed up after watching your video, however, it was a total fail for me. I’ll post my video soon and link to you when I’ve uploaded my results.
nice info, but what tools do you use to upscale the image?
www.bigjpg.com
@@thezinnystudio thanks
Great tips!
Great video Zinny. I'd like to know how much D-ID credits I'd need to make an average 15mins video
Hi, can you tell me what UPSCALER you use? Thanks
I just read through the comments and found it - thanks!
Alright
Very informative video. Thanks. I have a question. 1) You created "Sam" and made a couple of images. 2) Then, you created a new charactor named "Olivia" and made some images. 3) If you want to make more images of "Sam" the next day, will they keep the consistency as the day before?
Yes
Thank you so much this was very helpful. I also am creating a children's book. I have one question. Does this work for buildings and vehicles?
Thank you for your effort and tutorial, how do you keep character consistency like, if we open a new chat on GPT (let’s say the old chat was deleted), how do you continue to generate same character again in new chat with something like, GEN ID to recall the character?
Thank you for your time and looking forward to see you here.😊
I followed along and when I told it to design a 2D animation style, after a couple attempts, it gave me something perfect. I then told it so and that I wanted to see him in a different pose. The next result was a completely different design, like...dramatically different. I tried referencing the previous image, and it was still completely different again. I tried running it through an image describer and try and guide Consistent Character back to where I needed it but no matter what, it never generates anything the same.
Hi, my dear friend. Thank you for a complete video on how to make a character and talk. Like the character at the beginning of your video.
thank you
Thank you for this amazing tutorial. Can we perform this via API?
Do you mean, if you connect to Chatgpt via api?
This is a wonderful video, thanks for sharing. I learned many useful insights. Will have to sign for GPT Plus, I have been delaying the decision, but now... Great!
Glad it was helpful!
Wow this is exactly what I needed!
Hi Zinny, In this video at timestamp 10 mins : 56 seconds, You have upscaled the images. Could you please specify the site you have used to upscale and is it free for multiple images ? Waiting for your response. Thanks.
I use bigjpeg.com
So helpful thank you
uwc
Please which is the best between Chat Gpt and Bing AI for getting consistent characters for African folktales?
ChatGpt and Bing AI uses the same image model Dalle-3 so it doesnt matter which one you use.
@@thezinnystudio okay thanks ❤️
Awesome I subscribed.
Awesome, thank you!
Thank you for this video. Can I use mid journey to create consistent character for a story telling channel?
Yes you can!
@@thezinnystudio 💕
@@thezinnystudio I had to subscribe! Your creativity is amazing
Great video
Thanks!
Great tutorial! I have a question. How can I return to a character and its variations if I want or need more expressions or actions another day?
Yes you can!. just go back to that chat and continue prompting
@@thezinnystudio thank you, you are the best!
I'm not sure what I'm doing wrong, but it keeps changing from anime, to 2d illustration, to Pixar, to photography with each new image
Very helpful, thanks for sharing. I have a question, is that possible make these consistent images only in lines for colouring books?
I haven't tried that before, so can't say if it possible or not
Any tutorial on how to upscale the images?
I dont have a separate tutorial for upscaling. But is quite easy. Just visit bigjpeg.com. Select the images you want to upscale, chose how times you want to upscale, then upscale, download and use.
Something is wonky. When I try to download the images of my consistent character, they're downloading either in .xml or .webp Any clue? I tried asking the GPT and could never get it to download in an image format. I'm on my Lenovo Ideapad. Have never experienced a problem like this before.
Actually is not a problem. Chatgpt change the image format long time ago to webp. All you need to do is use a convert to convert to either jpeg or png.
@@thezinnystudio I finally found some place that talked about Google deciding they wanted smaller file sizes. I just pull them into Photoshop and change them now. Interesting though, I was working on a consistent character this afternoon. ChatGPT misunderstood my directions and created a landscape image with the character in different poses all on the same illio. I asked it to separate them and show them vertically, it did, and created them as PNGs. I had not expected that.
Would you suggest GPT4 or Midjourney as the best option for consistent characters?
It actually depends, the type of image you want. I use both but depending on the scenario
I’m following your instructions but mine is not coming out consistent. Did you have to train it first before you recorded this video?
Hmm, interesting. I didn't do any training before hand. As you can see in the tutorial, the images were generated live.
Try this to see if it works, back track to the last prompt before the no consistent image was created, edit the prompt and specify that image should be created with the last image generated without modification.
That way chatgpt will know you are referring to the very last image you created and want to continue using.
Great job 👏🏽
I notice in the there's an attachment symbol in the text box, does this mean we can create characters off of real pictures?
Yes and No. Yes in the sense that it will create an image using the picture as a reference. but it wont be exact replica that's the no part. But you can give it a try
@@thezinnystudio thank you!!
@@themommaspot5298 Uwc
thank you for this tutorial :D
You're welcome 😊
Thank you so much! Could I use this for an animal?
Yes of course!
If I upload a character I drew using the paperclip icon in chatgpt 4, will it show different poses?
can you show how I can generate multiple scene with the same set of characters. To do story books, we must mix multiple consistent characters. thanks
Will look into that
Hello lady! You are awesome! What upscaler do you use? Thank you
I currently use bigjpg.com. There are other free alternatives you can also try like Upscayl.
Great video and glad I could inspire you so much for the character consistency part. But I would appreciate, at least a shout-out next time please.
I have always had a shout out in my description linking your channel for pointing us to the GPT, it a huge time saver and thanks for that.
@@thezinnystudio . Ah yes. Sorry, as I'd watched the video and didn't hear it there and was just a bit surprised to see the very similar character prompts. No worries. Best of luck with the channel.
Thanks and all the best too.
Please what’s the art style or image style used for African folktales with this consistent ChatGPT ? Thanks
my images are not being consistant , the crown on my king changes every picture . what can i do to fix this ?
It's a great informative video,
is there any scope to create same consistent characters of entire family persons like 4 to 6 members in a family?
for a story telling video
I haven't tried to create more that two characters yet. But I will look into it
nice and Love this
Thanks
I like making these, they are so much fun.
thankyou for this . Greast Video. When i use this GPT it will only let me fdo 4 variations of the character, Is there any way to prevent this restriction?
I am not sure why is restricting you from generating more images. i will say, wait abit refresh the page and try again
Excellent straight to the point
Thank you
@thezinnystudio Great tutorial! Well explained and demonstrated. I am having trouble when requesting another expression. Consistent Characters GPT keeps giving me different characters and even races although I am specifically asking to use a particular image that it created. Please advise, and thank you in advance for your consideration regarding this matter
I noticed that too recently, I will get back to you, when I figure out how best to resolve it.
Not working for multi characters :
I'm here to focus on character designs and variations based on specific prompts about individual characters. If you're looking to illustrate scenes involving environments, interactions between characters, and detailed settings, I might not be able to fulfill this request directly. However, I can help create images based on descriptions of characters or their actions within a scene. Would you like to continue with character design or explore a different aspect of your story's visualization?
I hope you can help, I watched your video and then purchased a Chat GPT pro plan for $20 for a month. Now I am unable to access any Open API apps or get to the screen you are showing. Is this something about buying credits?
Hmm, that's strange, should have access to GPT store once you become a pro member. How are you accessing chatgpt, through your phone or a computer?
@@thezinnystudio I am using an HP laptop computer.
OK. Just wait abit, I am not sure if for new account if it takes time to show up. But I will research abit on it. You shouldn't have to do anything else.
@@thezinnystudio Thank you, I am currently in the help chat bot screen
Alright
Great advice, clearly concise. Can't wait to try this out!
Good luck!
What app do you use to make your avatar match what you are saying?
D-id
Is this done in the paid version ? Also how did you make ur mouth and eyes move
Yes it is done with paid version. and as for the talking avatar, check out this video ruclips.net/video/cqQl-PAsifg/видео.html