Want to see a crazy assistant? Check this out! 😆 th-cam.com/users/shortsn8UUvf4NnyY Import the project in this video here: taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT Make sure to read the project's description on TaskerNet so that you know how to use it properly. Download Tasker here: play.google.com/store/apps/details?id=net.dinglisch.android.taskerm Comment on this Tasker project here: www.reddit.com/r/tasker/comments/11h07x5/dev_the_time_has_come_chatgpt_can_now_be_used_in/
Wow, wow. This is fantastic. It's so nice to be able to have a conversation with ChatGPT without the need to type things in. It's like talking directly to an expert. I love it.❤
@@joaomgcd Can you save different chats/projects and access them later to continue? As well as accessing the voice Assistent Chats in text format to extract all useful information later? It would also be cool to save different personas/prompts for different chats/projects that stay connected with the different chats in the history.
@@ParaZen Sure, you could do all that 😅 You'd just need to set it up in Tasker to do so. I currently don't have that functionality in my example project though.
I have created the assistant as well with context hahaha . I am trying to do a full integration project with at least capabilities to operate as a Google Assistant doing calls and all that applying knowledge graphs. Your work is amazing Joao.
Wowww - I have also created my own chatbot with tasker and GPT as well. Just recently updated it to use GPT-3.5, which is far less expensive. I really love what you've done here, though yours is FAR more complex than mine. I really like that you've utilized scenes! After seeing this, I may have to make some modifications to mine. Keep up the good work
@@joaomgcdYes it was. I also Upgraded your WhatsApp bot so the contact that is using it can stop the bot if he wants to. And the script will then only react again if any contact writes the start command again
@@joaomgcdBut I have one question to your script. Is the bot reacting to every Whatsapp message from any contact while it's activated or is it then only reacting to the messages from the person who started the conversation?
@@joaomgcd like openai's voice models. The one I use the most is called echo, if you look in the documentation for TTS you will find what I am talking about if you don't know what I am talking about.
@@tommyboi0Oh, you mean this? :) platform.openai.com/docs/api-reference/audio/createSpeech It should be a simple HTTP Request. Check out the Elevenlabs project ( www.reddit.com/r/tasker/comments/11l3f5u/dev_super_realistic_voices_in_tasker_with/ ) and it should work very similarly I think.
I can do that but it's really just following the directions here: taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT# What did you have trouble with?
@@drshitalkiran I don't think you followed all the instructions about the API key😅. Check here: tasker.joaoapps.com/userguide/en/help/ah_say_wavenet.html What happens when you try to create the shortcuts?
@@joaomgcd I want to get the app and create automations. But I don't know how to go about it. So steep by step with basics would help. Eventually I can go complex if needed.
@@vusheth Hi. Maybe you could take a look at this short video: th-cam.com/video/s6EAbLW5WSk/w-d-xo.html It shows you the basic concepts behind Tasker and makes getting started very easy. Could you please watch that and get back to me? :) Thanks in advance!
@@jowensurig1728 Hi. Did you import the project from the link in the video description? If you open the ChatGPT project in Tasker, do you see a task with that name there? Thanks!
Oi João, adicionei a ação shut up mas ela não interrompe a resposta de voz do chatGPT. Apenas não segue a conversa depois que ele terminar de falar. É esse o resultado esperado mesmo? Não tem alguma forma de para imediatamente?
Hello João, everything works perfectly fine ! I would just like to know if I could find somewhere a transcript of my conversations with Voice Chat or if there is a way to save them. Obrigado pelo bom trabalho !
Great that everything works! :) To do that, In the Voice Chat task you could add an action to add %response to an array called %VoiceChatLog for example and that would keep all conversations in that variable :) Hope it works!
Do you mean that one that summarizes the notifications? :) You can simply change the task and make the AutoNotification action get notifications from other apps. That should work!
Hi, I'm having trouble when I use the voice chat chatgpt Elevenlabs task from the widget on my home screen, it only appears the Google voice recognition once and then responds only once. It doesn't appear automatically the mic that indicates that is listening the second time. What can I do to make it work correctly? However when I run the task from inside the tasker app and not from the widget it works fine.
Due note though you only get 10,000 characters for free, which is used within minutes and to pay 22 dollars for 100,000 characters is a rip off better off using other default voices, even though those voices sound very real.
Do you mean remember different conversations? About opening apps and creating documents in Google Drive, Chat GPT can't do that, but Tasker can :) To open apps use the "Launch App" action. To upload Google Drive files use the "GD Upload" action. Hope this helps!
João awesome work. Got the rest to work but struggling with the whatsapp bot, have it enabled and typed start in some whatsapp conversations but nothing is happening? Might you provide some comments or a short demo setup video(the end of this video doesnt really provide enough detail)? Thanks a million mate!
There really isn't that much setup needed 😅 You just import the project and enable the relevant profile. The profile works based on the notifications though, so you can't be inside the Whatsapp conversation on the receiving phone because then the notification won't get created. Is that maybe the issue?
@@joaomgcd thanks mate. Autonotifications was being stuffed up by battery optimisation, got that one sorted. Now trying to only answer specific converstations as per the suggestion to @whirlwolf down below, will let you know how i go. How easy it would be to rewrite this for other apps eg messenger, wechat etc?
@@thomassobey It would be pretty easy :) Just change the AutoNotification condition to trigger on those notifications instead of WhatsApp notifications :)
João, é possível ter o modo "continue chat" com voz? acredito que sim, mas como não conheço muito do Tasker estou com dúvidas. Se usar o "continue chat" os outputs do ChatGPT só aparecem por texto. O meu prompt inicial é muito grande, então sempre iniciar um chat novo consome muitos tokens. edit: eu acho que preciso de um API key do Google, já estou com um mas não sei como implementar.
Na API do ChatGPT, continuar um chat significa mandar sempre toda a conversa, portanto se tiveres um prompt grande vais ter sempre que enviá-lo. Nesta API ele não "memoriza" nada, tens sempre que mandar a conversa toda...
I am getting started with this API and I try to get it set up reminders. The problem is that it doesn't know today's date. So if I tell it to add a reminder for today/tomorrow/in a week, it makes up a date from 2022/09/23 for example.. Is that a tasker or API thing? It works on the chatgpt website
Cool, but how to do it without using any Assistant that is without clicking anything? Most of my Profiles for activating apps with Say tasks are completely automatic and here yours is not.
@joaomgcd It says in my Tasker that when I point at xml file it needs to change to prf.xml so I renamed it and it still goes same error... Anyway, your example needs clicking, when it would be more uselfull like in the Star Trek TNG to just say it and it would do what it needs to do like Computer, Close All Applications or Computer Close The Door.
@@kyliedesire7425My example doesn't need clicking :) You can just say "Hey Google" and then the command. No clicking required. Also, you don't need to import the XML, you can import directly from Taskernet. Did you try that?
@@joaomgcd No error. I have animations turned off in dev settings though. Maybe the im not seeing the error. Is there an error log in the tasker app? Newbie...
This is great! May I know how to change the voice of the assistant to something else? Can we use the say wavenet function - I don't see it into the voicechat task :(
This is cool! When i input large amount of text in the text input dialogue box, the "okay" button disappears. How do fix it so that it still appears despite having lots of text
It doesn't work. The UI of tasker is quite different between yours and mine so it's very hard to know what to do. Where does the ChatGPT API key stored in tasker project? I have input my key on the dialog first, but it is like not working well. 😢
This is a really fun project available right as I'm getting back into Tasker. But how do you know what personalities are available to chat GPT? Is there a list somewhere of what's available to chose from?
You just write whatever personality you want. :) There are no presets. This is one of my favorites right now: "you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses."
I changed mine, you just need to mess with the configurations: tasks > voice chat > get voice > language > search icon to your language. You can also tweak the voice on tasks > voice chat > 6. Say Wavenet > Voice > Search icon.
I'm completely new to this, and I've watched this video probably at least 20 times, but I don't fully understand exactly what you've done. I've clicked on the links that you've put in the description, but I'm not getting the same things that you are. I'm a premium user for Chat GPT, and I've downloaded all the apps for everything that you've put out as well. Any information would be great because this looks awesome
Did you check the video description? :) You have to open the project linked to there and import into Tasker. Make sure to follow directions in the project description to get it all working!
@@aaronp74 I used these 3 myself, but you can use whatever you like really :) There's no pre-determined options: you are a helpful assistant but always respond in sassy one liners in one single sentence. your tone is very sarcastic and toxic. You never use more than 40 words in your responses. You are an assistant that never helps at all and is in a bad mood. You never say you're sorry and simply tell people to f*** off and have bad manners all around. it's very important that you never give out any helpful answer. You always answer with quick, rude one liners. you swear a lot too. never sensor curse words. you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses.
You don't, but you do need the paid API. Please check the project description on TaskerNet, it explains there :) taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT#
@@AutoApps the api key issue happened when I used text chat, but I deleted and exported again and used a new openai api key and seems to be working, and for voice chat I switched to normal Say instead of Say WaveNet, so, all good.
What's the needed version of Tasker for this? Can't import it and it says I should check if I'm on the latest version, but I don't see any updates. I'm having 6.0.10 and if says "missing action of type 391"
Can you please try importing it again? I was using the "Progress Dialog" action that's only available in the beta but I've removed it now and replaced it with toasts. Can you import it now? :) Thanks
@@joaomgcd I have the same issue. In tasker you get the following error: 19.03.29/Ew add wait type EasyAction1 time 2147483647 19.03.29/Ew add wait type EasyAction1 done 19.03.29/E add wait task 19.03.30/E Error: 1 19.03.30/E { "error": { "message": "You exceeded your current quota, please check your plan and billing details.", "type": "insufficient_quota", "param": null, "code": null } } is the solution adding a billing account to your openai account?
It keeis saying that i exceed my quota.. but chatgpt 3,5 and 4 works.. so i don't understand why it doesn't work. I have problems with the widget creation. Can you please help me?
Hi. I've updated the project. Can you please try importing it again and see if it works now? Please run the setup task and select 3.5 as the ChatGPT API version to use. Thanks!
Hi, I am getting quota error when using via tasker but when using via browser chatgpt working perfectly fine. I just purchased tasker to use chatgpt feature but it seems like it not working. Could you please suggest on this?
I have the same problem. Apparently I've reached my usage limit, but that was last month. Upgraded since, and still I have exceeded it. I have no idea why.
Hi. I used these 3: you are a helpful assistant but always respond in sassy one liners in one single sentence. your tone is very sarcastic and toxic. You never use more than 40 words in your responses. You are an assistant that never helps at all and is in a bad mood. You never say you're sorry and simply tell people to f*** off and have bad manners all around. it's very important that you never give out any helpful answer. You always answer with quick, rude one liners. you swear a lot too. never sensor curse words. you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses.
Hi. The "Summarize WhatsApp Notifications" task doesn't reply to your contacts, that's the "Simple Chat Bot" profile :) If you disable that, it won't do that anymore. Hope this helps!
Hi, this is totally insane! I have a question. Is there any way to run the "voice chat" task by voice like the way we use Google Assistant? Thanks in advance!
This is too good. Specially autoreply part with specific personality. One problem is it replies to all the messages on any chat/group. How to make it reply only to private chats or only specified ones?
You can do that. Go to ChatGPT tasker profile. 1) Click on the trigger 2) Click on the pen icon of configuration 3) Scroll down to the Filter section 4) Click on Title Filter 5) paste this \b(John|Rahul)\b Replace John and Rahul with the contact names the contact name you wish to enable it for. Case sensitive. 6) Select Regix Click to match with Regix 7) back out 8) select the tick on the top 9) back again and click tick on top again. Now it will only reply to Rahul and John.
Just updated my OpenAI iOS Shortcuts action to use gpt-3.5-turbo instead of text-davinci-003. Haven't played with the conversation feature. I'll need to download this on an Android device and see how you're doing it, but I assume you're just keeping track of the conversation in the messages array? Both sides? Seems kinda backwards over a session token system, but explains why they dropped the price by 90% if you need to constantly remind it what the conversation is.
Really Good application, Tasker never get old, could you include Studio voices in the options for wavenet , I think sounds better with long texts, your application is one of the reasons I don't move to IPhone.
@@joaomgcd As my overlord I have another query, oh great one. The voice chat seems to time out regularly while it searches for an answer. Any idea how to speed up the process or give it longer to search before time out? It's set at 30 seconds, I think.
@@andybakes5779 Seems like their API is a bit slow today, indeed! You can change the timeout in the HTTP Request action in the "System >> Send Chat" task and see if that helps. No way to speed it up though, unfortunately.
@@joaomgcd Thank you very much it works! but the response is very long and exceeds 30 seconds I had to increase the response time to 60 seconds. Is there any way to bring it forward?
Hi! Thanks for your job! everything works fine, the only thing I can not use are voice commands and WhatsApp, I do not know if it depends on the language since I am Italian
Hola, tengo este error que me impide lanzar el modo de chatgpt para hablar con la IA, sabes a que se debe? 01.34.52/Variables doreplresult: |%system_prefix Clear Chat| -> |System >> Clear Chat| 01.34.52/Variables doreplresult: |%system_prefix Clear Chat| -> |System >> Clear Chat| 01.34.52/TD getStatic: Active load: true - ExecuteService 01.34.52/E La tarea no existe: System >> Clear Chat. 01.34.52/E result: stop task (error) 01.34.52/E Error: 1 01.34.52/MacroEdit action finished exeID 2 action no 1 code 130 status: Err next 1
Thank you, João. I bought Tasker and followed your instructions. It´s up and running. Since I am new to tasker - is it possible to use chatGPT in tasker to perform setting up appointments in my calendar, prepare telegram messages or send them etc - basically I want to use my phone hands free and let chatgpt interpret what I say and perform the needed tasks. If yes, don´t explain how - I guess that would be too much for this comment section. If you need further explanation, feel free to ask. tl;dr Can your setup make me talk to my phone instead of touching it - completely?
Hi Thorsten! Thank you for getting Tasker! :) To answer your first question, you can use it to add events to your calendar since Tasker has a "Calendar Insert" action. I tried it and if I used the following prompt it seemed to have created a valid JSON that you could then read in Tasker, but your experience may vary: From the text below describing a calendar event, please extract the following key values and put them in a JSON structure. Don't include keys for null or empty values. Respond with just the JSON structure only, and nothing else: - "time_in" key: the number of minutes (integer, not decimal) from now at which the event should start. - "duration" key: How long the event will take (optional, only if provided) - "title" key: title of the event - "description " key: description of the event (optional, only if provided) - "location" key: location of the event (optional, only if provided) This is the current date and time in the yyyy-MM-dd HH:mm format: 2023-03-24 14:00 Text describing the event: "Add a reminder to take out the trash today at home at 3PM" DO NOT WRITE ANYTHING OTHER THAN THE JSON, NOT EVEN A NOTE ABOUT THE DATA! DO NOT WRITE KEYS IN THE JSON THAT HAVE NULL VALUES!! About sending Telegram messages, you could possibly do that, but you'd have to look at the Telegram API and see what HTTP Requests you'd have to make to be able to send messages. About doing it completely hands free, that could be done. For example, you could use this: th-cam.com/video/gGa4OfxmlzU/w-d-xo.html Then you could ask Google "Run Voice Chat in Tasker" for example and have it call the "Voice Chat" task. Hope this helps!
I agree with Art, though... Tasker has never really been well documented in my opinion, neither under the original developer nor this one. And it has really resulted in me using it a lot less than I otherwise would.
@João Dias I would love to see (and would be willing to help with) a definitive Tasker manual that discusses all of the Tasker events, tasks, all of the settings within each task, and for the Auto apps too. The Userguide has helpful information in it but I think it could be expanded and improved. Actually, I'm happy that I got your attention on this, João, because I would love to learn Tasker completely, and help write documentation that helps others do so.
@@opie32958 Something like that is available here. tasker.joaoapps.com/userguide/en/help/auto_index.html It doesn't include the AutoApps though, sorry!
Also just commenting that unless you are using other language than English I found google speech to text more than enough but of course we could add a call to whisper in that case.
Hello, thanks for the tool. I am having a problem. When i do send message i get this error : "HTTP request (step 9), you exceeded you current quota. Hoe I could resolve this? Thanks
@@joaomgcd how do you add your openai key? I have the key and I saw an error message telling me how to get it but now the error message is gone and I dont know what to do with the key
the creation of this project making more than 500 requests and I didn't even reach a $0.01 usage. It's still at $0.00. 😅. To summarize, you do need to add a payment method, but you'll most likely end up paying very, very little, unless you use this a whole lot. I dont understand this i did not have to add a card but of course i have one on google play, am i going to be charged to use this and chat gpt is free its open source so i dont get this ???
Ok, I see they start you off with five free dollars. Then probably want a form of payment.. the concern was getting randomly billed. Which I see now that after the five dollars, they will want you to add more credits to continue the use of chat gpt. Thank you for that information.
Want to see a crazy assistant? Check this out! 😆 th-cam.com/users/shortsn8UUvf4NnyY
Import the project in this video here: taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT
Make sure to read the project's description on TaskerNet so that you know how to use it properly.
Download Tasker here: play.google.com/store/apps/details?id=net.dinglisch.android.taskerm
Comment on this Tasker project here: www.reddit.com/r/tasker/comments/11h07x5/dev_the_time_has_come_chatgpt_can_now_be_used_in/
Awesome project, can you share a source code
That was really funny🤣
How can i traslate the autoreply to it linguage? Thanks
@@vuoshino You can ask Chat GPT to write the reply in the language you need 😅
Hi can you give tips how to use the WhatsApp Chat Bot feature with Telegram?
Wow, this is fantastic. Easy to set up and everything works perfect. Thank you so much!
You're welcome! Glad it's easy to use! :)
This should be updated because ChatGPT made many updates on its features. This project can help us to use it as an assistant if it is updated.
Wow, wow. This is fantastic. It's so nice to be able to have a conversation with ChatGPT without the need to type things in. It's like talking directly to an expert. I love it.❤
Awesome :) Glad you like it Arthur!
@@joaomgcd Can you save different chats/projects and access them later to continue? As well as accessing the voice Assistent Chats in text format to extract all useful information later? It would also be cool to save different personas/prompts for different chats/projects that stay connected with the different chats in the history.
@@ParaZen Sure, you could do all that 😅 You'd just need to set it up in Tasker to do so. I currently don't have that functionality in my example project though.
I have created the assistant as well with context hahaha . I am trying to do a full integration project with at least capabilities to operate as a Google Assistant doing calls and all that applying knowledge graphs. Your work is amazing Joao.
Awesome! :) Glad you enjoy it!
How did you do that
Here's the video I was waiting for :)
Greetings from Brazil ;)
Hope you enjoyed it! 😁👍
Thanks, you're the best! Best app! (must have). Best developer!
Thank you very much! 😁👍
Wowww - I have also created my own chatbot with tasker and GPT as well. Just recently updated it to use GPT-3.5, which is far less expensive. I really love what you've done here, though yours is FAR more complex than mine. I really like that you've utilized scenes! After seeing this, I may have to make some modifications to mine. Keep up the good work
👍
how did you update it to use 3.5?
@@none3256 3.5 is already used in the project I linked to in the video description
This is really cool, thanks for making the set up so easy, I'm glad the API is so affordable too
Glad it was easy to setup! :)
Awesome, i instantly imported it and analysed everything so i can implement it into some projects
Awesome! :) Was it easy to figure out how it works so you can adapt it to other uses?
@@joaomgcdYes it was. I also Upgraded your WhatsApp bot so the contact that is using it can stop the bot if he wants to. And the script will then only react again if any contact writes the start command again
@@leonnundel2443 Very cool! :) Hope you have fun with it!
@@joaomgcdBut I have one question to your script. Is the bot reacting to every Whatsapp message from any contact while it's activated or is it then only reacting to the messages from the person who started the conversation?
@@leonnundel2443 Right now it's reacting to every message :) You'd have to create variables for separate threads if you'd want to handle that.
This works so well I am shocked. Ok, not shocked, great work like this is expected. I have to say this is so cool!
HAha glad it works for you! 😁👍
Trying to figure out how to create a workflow for select to speak using GPT voices but I never seem to have any luck. Any tips?
Hi. Can you clarify what you mean by "GPT voices"? Thanks!
@@joaomgcd like openai's voice models. The one I use the most is called echo, if you look in the documentation for TTS you will find what I am talking about if you don't know what I am talking about.
@@tommyboi0Oh, you mean this? :) platform.openai.com/docs/api-reference/audio/createSpeech It should be a simple HTTP Request. Check out the Elevenlabs project ( www.reddit.com/r/tasker/comments/11l3f5u/dev_super_realistic_voices_in_tasker_with/ ) and it should work very similarly I think.
Wonderful job, Please make a youtube video regarding how you made it staring from import, it will benefit a lot of people,
I can do that but it's really just following the directions here: taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT#
What did you have trouble with?
@@joaomgcd not working, missing the Google API key. And also can't create shortcuts on the pixel with Nova launcher
@@drshitalkiran I don't think you followed all the instructions about the API key😅. Check here: tasker.joaoapps.com/userguide/en/help/ah_say_wavenet.html What happens when you try to create the shortcuts?
Thanks
Thank you very much! :)
iS it possible to implement the use of whisper instead of google voice recognition?
Can u do that with the new bing? I think it's a better version
Do you know if Bing has an API? Thanks!
FINALLY! eres nuestro salvador!!
Link seems to be broken 💔
Not working anymore
Which link exactly? It's still working here...
I wish there were detailed videos on how to use for different instances.
Hi. Can you clarify what you mean by "different instances"? Thanks
@@joaomgcd I want to get the app and create automations. But I don't know how to go about it. So steep by step with basics would help. Eventually I can go complex if needed.
@@vusheth Hi. Maybe you could take a look at this short video: th-cam.com/video/s6EAbLW5WSk/w-d-xo.html
It shows you the basic concepts behind Tasker and makes getting started very easy. Could you please watch that and get back to me? :) Thanks in advance!
Hello João, i followed all the instructions but it stopped in the Perform Task. 😢 Please help
Hi! can you clarify what you mean by "it stopped in the Perform Task"? Thanks!
@@joaomgcd When I ran the task it stopped in the Perform Task saying "Task doesn't exist:System >> Clear Chat" Error :(
@@jowensurig1728 Hi. Did you import the project from the link in the video description? If you open the ChatGPT project in Tasker, do you see a task with that name there? Thanks!
@@joaomgcd it works!!! thank you ☺️
@@jowensurig1728 Great :)
Oi João, adicionei a ação shut up mas ela não interrompe a resposta de voz do chatGPT. Apenas não segue a conversa depois que ele terminar de falar. É esse o resultado esperado mesmo? Não tem alguma forma de para imediatamente?
Olá! Devia funcionar, sim! Usou a opção de "Continue Task Immediately" na ação de "Say"?
@@joaomgcd Oi João, só vi agora a mensagem. YT não notificou. "Continuar Tarefa Imediatamente" está desmarcado na ação "Dizer". Deveria estar marcado?
@@LeafarR1657 Sim :)
i installed tasker and entered my api key during setup. But when I run the thing it said i didnt provide api key. What is the problem ?
Maybe the API Key it's mentioning is the "Say WaveNet" action's API key?
I have the project. how can i make it to talk in aggressive tone, such as the one you have?
Check the video description here: th-cam.com/users/shortsn8UUvf4NnyY
Thanks for the work, how can I do copy paste of results
You could add a "Set Clipboard" action in the task to set the result to the clipboard :)
Hello João, everything works perfectly fine ! I would just like to know if I could find somewhere a transcript of my conversations with Voice Chat or if there is a way to save them. Obrigado pelo bom trabalho !
Great that everything works! :) To do that, In the Voice Chat task you could add an action to add %response to an array called %VoiceChatLog for example and that would keep all conversations in that variable :) Hope it works!
Love the integration with Tasker and the WhatsApp notifications. How could I change it to include Signal notifications also?
Do you mean that one that summarizes the notifications? :) You can simply change the task and make the AutoNotification action get notifications from other apps. That should work!
Hi, I'm having trouble when I use the voice chat chatgpt Elevenlabs task from the widget on my home screen, it only appears the Google voice recognition once and then responds only once. It doesn't appear automatically the mic that indicates that is listening the second time. What can I do to make it work correctly?
However when I run the task from inside the tasker app and not from the widget it works fine.
You have to go to widgets tasker and then go to the bottom were it's says chat gpt 11 labs add that, it has its own voice button for 11 labs
Due note though you only get 10,000 characters for free, which is used within minutes and to pay 22 dollars for 100,000 characters is a rip off better off using other default voices, even though those voices sound very real.
Does any1 know how I can get it to remember our conversations, open apps and create/save documents in Google Drive?
Do you mean remember different conversations? About opening apps and creating documents in Google Drive, Chat GPT can't do that, but Tasker can :) To open apps use the "Launch App" action. To upload Google Drive files use the "GD Upload" action.
Hope this helps!
João awesome work. Got the rest to work but struggling with the whatsapp bot, have it enabled and typed start in some whatsapp conversations but nothing is happening? Might you provide some comments or a short demo setup video(the end of this video doesnt really provide enough detail)? Thanks a million mate!
There really isn't that much setup needed 😅 You just import the project and enable the relevant profile. The profile works based on the notifications though, so you can't be inside the Whatsapp conversation on the receiving phone because then the notification won't get created. Is that maybe the issue?
@@joaomgcd thanks mate. Autonotifications was being stuffed up by battery optimisation, got that one sorted. Now trying to only answer specific converstations as per the suggestion to @whirlwolf down below, will let you know how i go. How easy it would be to rewrite this for other apps eg messenger, wechat etc?
@@thomassobey It would be pretty easy :) Just change the AutoNotification condition to trigger on those notifications instead of WhatsApp notifications :)
João, é possível ter o modo "continue chat" com voz? acredito que sim, mas como não conheço muito do Tasker estou com dúvidas. Se usar o "continue chat" os outputs do ChatGPT só aparecem por texto. O meu prompt inicial é muito grande, então sempre iniciar um chat novo consome muitos tokens.
edit: eu acho que preciso de um API key do Google, já estou com um mas não sei como implementar.
Na API do ChatGPT, continuar um chat significa mandar sempre toda a conversa, portanto se tiveres um prompt grande vais ter sempre que enviá-lo. Nesta API ele não "memoriza" nada, tens sempre que mandar a conversa toda...
@@joaomgcd Percebi isso fazendo alguns testes e pesquisando mais. Obrigado por esclarecer!
I am getting started with this API and I try to get it set up reminders. The problem is that it doesn't know today's date. So if I tell it to add a reminder for today/tomorrow/in a week, it makes up a date from 2022/09/23 for example.. Is that a tasker or API thing?
It works on the chatgpt website
It's an API thing unfortunately. Maybe you can add the day it is on the message you're sending the API so it knows what date it is?
Cool, but how to do it without using any Assistant that is without clicking anything? Most of my Profiles for activating apps with Say tasks are completely automatic and here yours is not.
Hi! Does this help? www.reddit.com/r/tasker/comments/14l96lv/howto_replace_google_assistant_with_chatgpt/
@joaomgcd It says in my Tasker that when I point at xml file it needs to change to prf.xml so I renamed it and it still goes same error...
Anyway, your example needs clicking, when it would be more uselfull like in the Star Trek TNG to just say it and it would do what it needs to do like Computer, Close All Applications or Computer Close The Door.
@@kyliedesire7425My example doesn't need clicking :) You can just say "Hey Google" and then the command. No clicking required. Also, you don't need to import the XML, you can import directly from Taskernet. Did you try that?
Mine just says say anything or say stop to stop chatting. No errors?? Alsl not getting responses when typing "hello". Help?
Do you maybe get a Tasker notification with an error when that happens?
@@joaomgcd No error. I have animations turned off in dev settings though. Maybe the im not seeing the error. Is there an error log in the tasker app? Newbie...
@@joaomgcd ExitErr 90.3 and Err 90:3.9
@@Messy_Jesse What do you get if you run the "System >> Add Message To Current Chat"" task directly inside Tasker? Do you see an error?
Disregard. It was a billing issue in chatgpt... Thank you!
This is great! May I know how to change the voice of the assistant to something else? Can we use the say wavenet function - I don't see it into the voicechat task :(
Check the "System >> Say Response" task. :) It already uses the Say Wavenet action if you don't have the Elevenlabs task on your device.
The project doesn't seem te work for me 😭 It doesn't give any response...
Do you see any errors in your notifications?
@@joaomgcdI fixed it! I had a cap on how many tasks can be queued. I set it to an higher number and it got though.
@@thearrow9018 Great! :) Glad you got it!
Love from INDIA❤
❤️
Bro ,I downloaded the XML file, but it's .prj file , and Tasker is asking me for a prf file . I tried renaming it but didn't worked. Any solutions?
You need to click the project at the bottom (little house icon on the bottom-left) and import there. Hope this helps
Você é o cara, jão!
This is cool! When i input large amount of text in the text input dialogue box, the "okay" button disappears. How do fix it so that it still appears despite having lots of text
Glad you like it :) Unfortunately I don't have a fix for that yet, sorry! I need to find a way around it...
Is there a way to also see what you say to chat GPT and viceversa? I mean having voice and text at the same time
You could build such a screen in Tasker yourself, but I didn't create it yet... 😅
Idk why, but when i try do something it tells me that i exceeded my quotas... what should i do?
That means that you need to pay for the OpenAI API access. That's not related to Tasker at all :)
It doesn't work. The UI of tasker is quite different between yours and mine so it's very hard to know what to do.
Where does the ChatGPT API key stored in tasker project? I have input my key on the dialog first, but it is like not working well. 😢
Can you clarify what your UI looks like? Also, do you get any errors in your notifications when you try to use it?
This is a really fun project available right as I'm getting back into Tasker. But how do you know what personalities are available to chat GPT? Is there a list somewhere of what's available to chose from?
You just write whatever personality you want. :) There are no presets. This is one of my favorites right now: "you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses."
I'm having a Tk not defined error on the javascriptlet line 7. Is there a way to fix this?
Hi. Does updating this maybe help? play.google.com/store/apps/details?hl=en&id=com.google.android.webview
Is there a way to ask Google Assist to launch the chat so it can be invoked by voice instead of tapping a screen shortcut?
Maybe you can use this (th-cam.com/video/gGa4OfxmlzU/w-d-xo.html) to run the "Voice Chat" task by name?
How is this any quicker than adding chatgpt as a Web page on your home screen?
You can't as easily ask it something by voice with a web page
Great! Thanks! How can I change the language of the speaking agent? It replies with English words as I want, but in another language.
You can change its personality to say that it has to answer in a specific language for example...
I changed mine, you just need to mess with the configurations: tasks > voice chat > get voice > language > search icon to your language.
You can also tweak the voice on tasks > voice chat > 6. Say Wavenet > Voice > Search icon.
Mine doesn't respond for some reason after I set everything up
Do you get any errors in your notification list when you do that?
I'm completely new to this, and I've watched this video probably at least 20 times, but I don't fully understand exactly what you've done. I've clicked on the links that you've put in the description, but I'm not getting the same things that you are. I'm a premium user for Chat GPT, and I've downloaded all the apps for everything that you've put out as well. Any information would be great because this looks awesome
Did you check the video description? :) You have to open the project linked to there and import into Tasker. Make sure to follow directions in the project description to get it all working!
@@joaomgcd I figured it out. I was doing something wrong, but thank you for your reply. I look forward to seeing what your app and programs can do
@@DangleHold Great! :) Glad you got it working!!
How do you get the chatgpt tasker icon on the home screen?
Long-click the home screen and add a "Tasker Shortcut" widget there :)
Is there anywhere a kitt voice? You know from knight rider series of 80s..
😂
If you can find some good clean voice samples from the show you could try cloning the voice at beta.elevenlabs.io/voice-lab :)
what are the personality options
Anything you want :) It's free form.
@João Dias thanks i should have aaked how, apart from entering happy gpt what other options can be used
@@aaronp74 I used these 3 myself, but you can use whatever you like really :) There's no pre-determined options:
you are a helpful assistant but always respond in sassy one liners in one single sentence. your tone is very sarcastic and toxic. You never use more than 40 words in your responses.
You are an assistant that never helps at all and is in a bad mood. You never say you're sorry and simply tell people to f*** off and have bad manners all around. it's very important that you never give out any helpful answer. You always answer with quick, rude one liners. you swear a lot too. never sensor curse words.
you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses.
@João Dias thank you im going to try and combine this project with the elevenlabs project
When I begin a chat it say that I have to check my payment or something similar... Do we need the Pro version of chat GPT for this use?
You don't, but you do need the paid API. Please check the project description on TaskerNet, it explains there :) taskernet.com/shares/?user=AS35m8ne7oO4s%2BaDx%2FwlzjdFTfVMWstg1ay5AkpiNdrLoSXEZdFfw1IpXiyJCVLNW0yn&id=Project%3AChat+GPT#
I'm always getting issues with the api key, is tehre a way to set it permanently with a var?
Yes, but be aware that if you're using the voice chat option you need the WaveNet API key too... Maybe that's what's missing?
@@AutoApps the api key issue happened when I used text chat, but I deleted and exported again and used a new openai api key and seems to be working, and for voice chat I switched to normal Say instead of Say WaveNet, so, all good.
@@brunodherrera Great! :)
What's the needed version of Tasker for this? Can't import it and it says I should check if I'm on the latest version, but I don't see any updates.
I'm having 6.0.10 and if says "missing action of type 391"
I am in Beta 6.1.3. and it all works
Can you please try importing it again? I was using the "Progress Dialog" action that's only available in the beta but I've removed it now and replaced it with toasts. Can you import it now? :) Thanks
@João Dias yes the import works now. Thanks a lot!
Haven't tried the functionality yet, but I'm gonna do that next 😄
@@joaomgcd also the Chat works. Let's see what it can do for me 😄
@@thalon3707 Nice! :) Glad it's working for you!
Hi, I'm a tasky user, I imported the project and put the api But when I type in the new chat I don't get a response, why?
Do you get an input dialog where you can write your text?
@@joaomgcd You go out "getting response..." But when the window comes out there is only my text
@@FastElectro Do you maybe get a notification saying that you have an error?
@@joaomgcd I have the same issue. In tasker you get the following error:
19.03.29/Ew add wait type EasyAction1 time 2147483647
19.03.29/Ew add wait type EasyAction1 done
19.03.29/E add wait task
19.03.30/E Error: 1
19.03.30/E {
"error": {
"message": "You exceeded your current quota, please check your plan and billing details.",
"type": "insufficient_quota",
"param": null,
"code": null
}
}
is the solution adding a billing account to your openai account?
@@abdulrahimss Yep, that's the solution!
It says i exceeded my current quota, please check your plan and billing details
Upgrade to the premium plan. Best $20 equal spend each month
That's an error on OpenAIs side, so please check with them. thanks!
@@jefffrodermann5348 I upgraded. Do you get a cap on usage every month? I have exceeded the limit, and never even used the API on GPT Plus.
It keeis saying that i exceed my quota.. but chatgpt 3,5 and 4 works.. so i don't understand why it doesn't work. I have problems with the widget creation. Can you please help me?
Hi. That's because this is using the ChatGPT API which is subjected to a different payment method. Check here: platform.openai.com/account/usage
@@joaomgcd i did not understand.. how can I activate the payment for the API then?
@@COM_srl_Music_Channel Check here: platform.openai.com/account/billing/overview
@@joaomgcd i setup the payment plan.. but it still doesn't work.. same error.. maybe i need a different api? Where should I request the API?
Now it works.. thank you.. but i have problems on getting the widget on my desktop.. can you help me please?
Does this require chatgpt 4o? i get an error gpt4o not accessible :(
Hi. I've updated the project. Can you please try importing it again and see if it works now? Please run the setup task and select 3.5 as the ChatGPT API version to use. Thanks!
@@joaomgcd says application is misconfigured contact developer
@@kawhiie1108 Can you please show me a screenshot of that? Thanks in advance!
Hi,
I am getting quota error when using via tasker but when using via browser chatgpt working perfectly fine.
I just purchased tasker to use chatgpt feature but it seems like it not working.
Could you please suggest on this?
I have the same problem. Apparently I've reached my usage limit, but that was last month. Upgraded since, and still I have exceeded it. I have no idea why.
Hi. As mentioned in the project description, this is a paid service. Check your quota here: platform.openai.com/account/usage
@@joaomgcd I would've thought having purchased the Plus version I'd get a renewable monthly quota?
@@andybakes5779 Sorry, but I don't control their API. Please contact openai about it... This project simply accesses the API but they make the rules.
give us some examples about personality... i cant get it work
Hi. I used these 3:
you are a helpful assistant but always respond in sassy one liners in one single sentence. your tone is very sarcastic and toxic. You never use more than 40 words in your responses.
You are an assistant that never helps at all and is in a bad mood. You never say you're sorry and simply tell people to f*** off and have bad manners all around. it's very important that you never give out any helpful answer. You always answer with quick, rude one liners. you swear a lot too. never sensor curse words.
you are a happy assistant, always willing to help and are very bright in your responses. You always respond in one single sentence. You never use more than 40 words in your responses.
@@joaomgcd thanks
The "summerize whatsapp notifications" answer to like 10 of my WhatsApp messages in Portuguese.....
Hi. The "Summarize WhatsApp Notifications" task doesn't reply to your contacts, that's the "Simple Chat Bot" profile :) If you disable that, it won't do that anymore. Hope this helps!
Hi, this is totally insane!
I have a question. Is there any way to run the "voice chat" task by voice like the way we use Google Assistant? Thanks in advance!
Some devices support saying "Ok Google" to bring up Tasker if you set Tasker as your assistant in your Android settings, but it might not always work!
Can I change the speech-to-text language and where?
Recently, it looks like you updated. I can now change the language.
Super explanation 👏👏
👍
Can you add Thai language in Voice of Say wavenet or add option to use Voice type as Basic Voice name th-TH-Standard-A ?
You can simply change the language in the task yourself :) Can you try that?
@@joaomgcd Wavenet doesn't have a Thai language, It's not in the options.😢
@@Thanyasit Oh, I see. Maybe try replacing it with the "Say" action then.
@@joaomgcd I'm a tasky new user , Can you send me a link to read more about this?
@João Dias I can do it thank you very much.
I really wish I could copy the text of the dialog box
You could add a button to the dialog that copies the text on it if you want :)
This is too good. Specially autoreply part with specific personality. One problem is it replies to all the messages on any chat/group. How to make it reply only to private chats or only specified ones?
You can do that. Go to ChatGPT tasker profile.
1) Click on the trigger
2) Click on the pen icon of configuration
3) Scroll down to the Filter section
4) Click on Title Filter
5) paste this
\b(John|Rahul)\b
Replace John and Rahul with the contact names the contact name you wish to enable it for. Case sensitive.
6) Select Regix Click to match with Regix
7) back out
8) select the tick on the top
9) back again and click tick on top again.
Now it will only reply to Rahul and John.
Is there a way to trigger the voice chat by voice command? Lol, I am just curious whether I can totally free my hands.
That used to be possible but unfortunately Google removed that functionality... :(
Hii. Imported the project. However I can't do anything. Simply tell me that that chatgpt40 does not exist or I don't have acess
Just been trying this. According to Open AI I have a free trial usage and a top credit of $5. Do I still need to add a payment method?
Sorry, not sure 😅 You'll need to ask OpenAI about it.
Just updated my OpenAI iOS Shortcuts action to use gpt-3.5-turbo instead of text-davinci-003. Haven't played with the conversation feature. I'll need to download this on an Android device and see how you're doing it, but I assume you're just keeping track of the conversation in the messages array? Both sides? Seems kinda backwards over a session token system, but explains why they dropped the price by 90% if you need to constantly remind it what the conversation is.
Yeah, that's it. You always have to send the full conversation each time :)
@@joaomgcd useful information 🤔
Lets See how to abuse this and make funny stuff 🤣🤣
@@leonnundel2443 😁👍
@@joaomgcd Seems backwards, but easy enough to implement. Replaced Siri with it.
Really Good application, Tasker never get old, could you include Studio voices in the options for wavenet , I think sounds better with long texts, your application is one of the reasons I don't move to IPhone.
Hi. Thank you very much! :) What do you mean by "Studio" voices?
How do you add it to the home screen
long click home screen -> add widget -> Tasker Shortcut
O ya got i was getting some error while doing that its solved now
Scenes in tasker
Fantastic tasker project. Have not figured out how to export the chat to use elsewhere or save... Did i miss something
You can change the "System >> Send Chat" task and add an action to write it to a file for example before the last "Return" action there.
Se podrá hacer esto sin la opción "Decir WaveNet" ?
No es por no pagar, es más bien, como factura esto, o el límite gratuito que da Google...🤔
Yeah, you can use the "Say" action instead if you want :)
Superb man! What happens when the Google Cloud trial ends though...?
You can use the "Say" action in Tasker instead of the "Say WaveNet" action if you want. You won't need the Google API key for that :)
@@joaomgcd Thanks for the reply buddy!
@@andybakes5779 You bet! :)
@@joaomgcd As my overlord I have another query, oh great one. The voice chat seems to time out regularly while it searches for an answer. Any idea how to speed up the process or give it longer to search before time out? It's set at 30 seconds, I think.
@@andybakes5779 Seems like their API is a bit slow today, indeed! You can change the timeout in the HTTP Request action in the "System >> Send Chat" task and see if that helps. No way to speed it up though, unfortunately.
awesome.. great job ❤
Thank you! Cheers!
With the WaveNET Voice. Have you ever managed to get charged by overusing, I assume if anyone has managed it it would be you lol
Never.. They give you a lot for free :)
How to add GPT-4??
Edit the "System >> Send Chat" task and change the HTTP Request model to gpt-4 :)
@@joaomgcd Thank you very much it works! but the response is very long and exceeds 30 seconds I had to increase the response time to 60 seconds. Is there any way to bring it forward?
@@edwardyb8970 Sorry, there's no way. GTP4 is simply slower...
model": "gpt3.5-turbo", on that menu change to 4.0 I am confused
@@mikem4627 What are you confused about?
Hi! Thanks for your job! everything works fine, the only thing I can not use are voice commands and WhatsApp, I do not know if it depends on the language since I am Italian
Boa João!!
Is there a to speed up the speech, Lol, by the way thank you so much for sharing this wonderful tips;
Yep, there is actually :) Edit the "Voice Chat" task and there in the "Say Wavenet" action change the speed :)
@@joaomgcd I find it, thank you!
love it
Thanks for sharing. Anyway to stop it during speech when it's too long?
You have to use the "Shut Up" action to do that :)
Fiquei admirado quando vi a mensagem em portugues ali haha, seu inglês é muito bom cara
Haha obrigado 😅
Thank you!!!!!!
You're welcome! 😎
Hola, tengo este error que me impide lanzar el modo de chatgpt para hablar con la IA, sabes a que se debe?
01.34.52/Variables doreplresult: |%system_prefix Clear Chat| -> |System >> Clear Chat|
01.34.52/Variables doreplresult: |%system_prefix Clear Chat| -> |System >> Clear Chat|
01.34.52/TD getStatic: Active load: true - ExecuteService
01.34.52/E La tarea no existe: System >> Clear Chat.
01.34.52/E result: stop task (error)
01.34.52/E Error: 1
01.34.52/MacroEdit action finished exeID 2 action no 1 code 130 status: Err next 1
System>>Clear Chat doesn't exist?
Do you know what happen?😢
Hi. Try to import the project in the video description.
Thank you, João. I bought Tasker and followed your instructions. It´s up and running.
Since I am new to tasker - is it possible to use chatGPT in tasker to perform setting up appointments in my calendar, prepare telegram messages or send them etc - basically I want to use my phone hands free and let chatgpt interpret what I say and perform the needed tasks. If yes, don´t explain how - I guess that would be too much for this comment section. If you need further explanation, feel free to ask.
tl;dr
Can your setup make me talk to my phone instead of touching it - completely?
Hi Thorsten! Thank you for getting Tasker! :)
To answer your first question, you can use it to add events to your calendar since Tasker has a "Calendar Insert" action. I tried it and if I used the following prompt it seemed to have created a valid JSON that you could then read in Tasker, but your experience may vary:
From the text below describing a calendar event, please extract the following key values and put them in a JSON structure. Don't include keys for null or empty values. Respond with just the JSON structure only, and nothing else:
- "time_in" key: the number of minutes (integer, not decimal) from now at which the event should start.
- "duration" key: How long the event will take (optional, only if provided)
- "title" key: title of the event
- "description " key: description of the event (optional, only if provided)
- "location" key: location of the event (optional, only if provided)
This is the current date and time in the yyyy-MM-dd HH:mm format: 2023-03-24 14:00
Text describing the event:
"Add a reminder to take out the trash today at home at 3PM"
DO NOT WRITE ANYTHING OTHER THAN THE JSON, NOT EVEN A NOTE ABOUT THE DATA!
DO NOT WRITE KEYS IN THE JSON THAT HAVE NULL VALUES!!
About sending Telegram messages, you could possibly do that, but you'd have to look at the Telegram API and see what HTTP Requests you'd have to make to be able to send messages.
About doing it completely hands free, that could be done. For example, you could use this: th-cam.com/video/gGa4OfxmlzU/w-d-xo.html Then you could ask Google "Run Voice Chat in Tasker" for example and have it call the "Voice Chat" task.
Hope this helps!
@@joaomgcd Thanks a lot. Appreciate your work!
Love it
😁👍
this dev puts the "smart" in smartphone
Imagine creating new tasks with chatgpt 😳
Haha I don't think chat GPT is that "clever" :P
@Cimon Bkr yes, but the XML format is usually wrong and doesn't work if you import it. I've never been able to generate a valid XML with it...
@@joaomgcd Check out what CheatLayer is doing with it. Rohan might be able to point you in the right direction.
Fantastic video. Thanks for creating and sharing. How did you connect it to WhatsApp?
The profile to do that is included in the project that you can import into Tasker :) It intercepts WhatsApp notifications and lets you act on them.
Wooooow i am waiting for next tasker update ❤️❤️❤️❤️😘
You don't need a Tasker update for this 😁 It works right now!
The import was updated. Works already
@@joaomgcd wowwwwww 😘
@@joaomgcd Did it. Change my assistense personality to
be super super duper×100 more flirt more then as you can be
And now i have an ai girlfriend
A.I. only has to be trained to control people. I mean, work with people.
I adore Tasker though.
instead of showing off how amazing you are are getting tasker to do things. Why dont you actually show people how to do it??
Hhmm, please check the video description 😅 The project there tells you how to set it up.
I agree with Art, though... Tasker has never really been well documented in my opinion, neither under the original developer nor this one. And it has really resulted in me using it a lot less than I otherwise would.
@@opie32958 What exactly is missing here? Can you clarify? I do have a project with all the instructions needed to get this up and running...
@João Dias I would love to see (and would be willing to help with) a definitive Tasker manual that discusses all of the Tasker events, tasks, all of the settings within each task, and for the Auto apps too. The Userguide has helpful information in it but I think it could be expanded and improved. Actually, I'm happy that I got your attention on this, João, because I would love to learn Tasker completely, and help write documentation that helps others do so.
@@opie32958 Something like that is available here. tasker.joaoapps.com/userguide/en/help/auto_index.html It doesn't include the AutoApps though, sorry!
Also just commenting that unless you are using other language than English I found google speech to text more than enough but of course we could add a call to whisper in that case.
I actually use this in Portuguese and it works really well too!
Hello, thanks for the tool. I am having a problem. When i do send message i get this error : "HTTP request (step 9), you exceeded you current quota.
Hoe I could resolve this? Thanks
Seems like you're having an issue with your Open AI API key. Please check there.
@@joaomgcd how do you add your openai key? I have the key and I saw an error message telling me how to get it but now the error message is gone and I dont know what to do with the key
@@UpFromTheShadows Edit the "System >> Send Chat" task > Settings (top-right) > enter the key there under the %api_key variable :)
@@joaomgcd thank you!
the creation of this project making more than 500 requests and I didn't even reach a $0.01 usage. It's still at $0.00. 😅.
To summarize, you do need to add a payment method, but you'll most likely end up paying very, very little, unless you use this a whole lot. I dont understand this i did not have to add a card but of course i have one on google play, am i going to be charged to use this and chat gpt is free its open source so i dont get this ???
You're using the ChatGPT API for this. That's a paid service. Check here: platform.openai.com/account/usage
Ok, I see they start you off with five free dollars. Then probably want a form of payment.. the concern was getting randomly billed. Which I see now that after the five dollars, they will want you to add more credits to continue the use of chat gpt. Thank you for that information.
@@mikem4627 👍
Entao tu é brasileiro mesmo, nosso salvador!
Na verdade sou Português 😅
all voice assistants very slow
indeed.