Loved this tutorial! I would love to see one about connecting chat GPT to WordPress for blogging. Especially when it comes to creating long form articles with multiple OpenAI modules, getting consistent content across those modules, and uploading with markdown format in WordPress. :)
What a request! That would probably be a 30-45 minute tutorial. This gives us some ideas though. The multiple OpenAi modules specifically is a great call out. Well put something similar in the queue 😎
Thank you for this valuable video. How on the other hand to remove stray words where GPT incorporates context into its response. And how to ask GPT to make on the same conversation 3 answers. And compile these 3 answers.
Thank you! Great questions: Re Stray words: nothing you can do about this. Try to adjust your prompt to include an “example” response. That might help tighten the answer GPT gives you Re 3 answers: you’ll need to create 3 modules that ask the same question Re compile them: after the 3 modules above respond, you could assign them to a single variable in Make, or just input all 3 responses into (wherever) you wanted to output the GPT responses. Hope that helps!
@@XRayAutomation I was able to make it work with the openai module, it was awesome! but after 4 "runs" i kept getting response that "As an ai model i don't have real time internet acccess" even though i never changed the workflow. Is this something i can get your help with?
Is it possible to create a voice bot in telegram, to which I will send an audio message. After the bot will transcribe and perform tasks based on the message. For example, to change an event in the calendar
Great question... Based on Telegrams support for Make (xray.tools/app/telegram) -- I'd say... kinda. What you're describing is tremendously complicated because the automation would be quite long, and need access to a huge array of your systems (such as your calendar). It would also be very brittle. If your phrasing or the transcription doesn't match the exact key-words you configure, it wouldn't work as you intend. We're still quite far away nocoders being able to 'roll your own' assistant. For now, sadly, I'd say stick with Siri or Google Assistant.
It should just be the response variable from OpenAi. However, depending on the app you're inserting the response in - that may be eliminating spaces & formatting.
This is a bit INCOMPLETE! Why do all Vids like this Omit the OAUTH procedure steps before you can use a standard Gmail account. Please do the above cleanly and clearly. As the only step by step vid is an Outdated vid referencing Integromat (aka Make) Since Make have NOT updated their video it introduces Errors! during the interface to the Gmail module. Your good video here? literally is non effective. Without an updated OAuth process. Will subscribe if and when you create the OAuth process! In the meantime, you got a comment to sweeten the YT algorithm.
Thank you for the comment, and thank you for watching! You're right theres a BIG difference between (standard) Gmail and your (company) Gmail account. If you're still searching @spicer41282 -- We made a video to directly address this comment. Hope it helps! th-cam.com/video/d0u7IYQJXDM/w-d-xo.html
This was by far the best tutorial. Thank you
You are very welcome! Tell a friend, and let us know what else you would like to learn about!
Exactly !❤
Thank you for this, I have also added a "Router" to send it to my Telegram & Discord.
Now you’re getting fancy 😎
great tutorial.
僕は日本人です！とても興味深く見てます！よかったらもっと色々な活用法を見たいです！
ご視聴いただきありがとうございます！字幕はお使いですか？何か質問はありますか？
学んでいただけて嬉しいです！
-ChatGPT helped us here 😅
Loved this tutorial! I would love to see one about connecting chat GPT to WordPress for blogging. Especially when it comes to creating long form articles with multiple OpenAI modules, getting consistent content across those modules, and uploading with markdown format in WordPress. :)
What a request!
That would probably be a 30-45 minute tutorial. This gives us some ideas though. The multiple OpenAi modules specifically is a great call out.
Well put something similar in the queue 😎
Thanks!
Thank you!
Thank you for this valuable video. How on the other hand to remove stray words where GPT incorporates context into its response. And how to ask GPT to make on the same conversation 3 answers. And compile these 3 answers.
Thank you! Great questions:
Re Stray words: nothing you can do about this. Try to adjust your prompt to include an “example” response. That might help tighten the answer GPT gives you
Re 3 answers: you’ll need to create 3 modules that ask the same question
Re compile them: after the 3 modules above respond, you could assign them to a single variable in Make, or just input all 3 responses into (wherever) you wanted to output the GPT responses.
Hope that helps!
How do you get it to properly format the email rather than having it show up as one undifferentiated blob (paragraph)?
Sounds like you're missing a Mailto format: Check this wikipedia page.
en.wikipedia.org/wiki/Mailto
This guy is a Live Pixar Character!! 🤔😬
Is it possible to run prompts on Make that will use web browsing ?
Yup! Try the OpenAI module or HTTP
@@XRayAutomation I was able to make it work with the openai module, it was awesome! but after 4 "runs" i kept getting response that "As an ai model i don't have real time internet acccess" even though i never changed the workflow. Is this something i can get your help with?
Book some time at lowcodeengineers.com and we can help!
Is it possible to create a voice bot in telegram, to which I will send an audio message. After the bot will transcribe and perform tasks based on the message. For example, to change an event in the calendar
Great question... Based on Telegrams support for Make (xray.tools/app/telegram) -- I'd say... kinda.
What you're describing is tremendously complicated because the automation would be quite long, and need access to a huge array of your systems (such as your calendar). It would also be very brittle. If your phrasing or the transcription doesn't match the exact key-words you configure, it wouldn't work as you intend.
We're still quite far away nocoders being able to 'roll your own' assistant. For now, sadly, I'd say stick with Siri or Google Assistant.
awesome. I'm trying to do this with a customGPT, is my intuition correct in me needing to use HTTP module instead of the openAi module?
You could use either one! Just depends on what exactly you’re trying to do!
Nice video. But can't I use my botpress variables for the chatgpt prompt.
You totally could!
Botpress is still really “new” on the low-code scene. We’ll make a video for that sometime soon too!
Is this affected by having a plus account or is everything still the same?
You’ll have access to GPT4!
Subscribed
- checkout www.XRay.Tools
- build smarter 🧠
- build faster 💪
How can I keep the exact GPT message format, including Carriage returns and live feed?
It should just be the response variable from OpenAi.
However, depending on the app you're inserting the response in - that may be eliminating spaces & formatting.
How do you fit the email formatting? Great video, thanks!
Try using a different gpt model if you’re having token limits because an email is too long.
Glad you liked it!
This is a bit INCOMPLETE!
Why do all Vids like this Omit the OAUTH procedure steps before you can use a standard Gmail account.
Please do the above cleanly and clearly. As the only step by step vid is an Outdated vid referencing Integromat (aka Make)
Since Make have NOT updated their video it introduces Errors! during the interface to the Gmail module.
Your good video here? literally is non effective.
Without an updated OAuth process.
Will subscribe if and when you create the OAuth process!
In the meantime, you got a comment to sweeten the YT algorithm.
Thank you for the comment, and thank you for watching!
You're right theres a BIG difference between (standard) Gmail and your (company) Gmail account.
If you're still searching @spicer41282 -- We made a video to directly address this comment. Hope it helps!
th-cam.com/video/d0u7IYQJXDM/w-d-xo.html