วีดีโอ

Thanks ! <3

You need to know the Linkedin identifier of that person and write it in a specific format on make.com. This is an example: Hello, my name is @[Jane Doe](urn:li:person:123456). I work in @[Company Name](urn:li:organization:123456). You can join my discord and people will help you there ;)

For sure ;)

Thanks ! I guess you need to use the module "watch changes", set up a make.com webhook and make sure you select the E2 cell in your range option inside the Make.com add-on in google sheets. The make.com scenario can then call perplexity to what you want. If you're not sure what this means, you can join the discord server ;)

Send me an email at hello@neurafirst.com

Hello, I think you need 2 loops in your make.com blueprint 1 that iterate over all rows of your google sheet, and a 2nd loop that go over all columns in the same row to add up the amounts in the right cells and update the total cell

I just double checked, you can find the headers info at 12:37 ;)

Hmm that's weird, what happens when you try to create the connection to google sheets ? Normally it should prompt you to your google email account, and then you should allow Make.com to connect to Google sheets.

not as a podcast, but rather as a normal business users

You could upload your data in NotebookLM and then ask it to write an article in the form of a blog post ( or a Linkedin post for example )

I replied to you in the discord server, I think it can easily be fixed ;)

Thanks ! If you can't make it work, you can always join the discord and ask for help. Plenty of knowledgeable people there ;)

You're welcome ! ;)

You're welcome ! ;)

Well it depends on the length of the text haha. You could create a " short takeaway podcast" out of a very long article.

@@Smail-AI and if you read the transcript of that podcast, it would take much less time.... Nothing slower than listening to podcasts, is there?

@@michaelbliss3337 I personally listen to podcasts before sleeping or while cooking. So I'm not even considering speed as an important factor haha

Thanks ! What kind of other content would you like to see next ?

Sure I can do that. I added this to my list ;)

Thank you !

Haha thanks I also love disco/funk music ;)

Thank you so much ! :)

Yeah it is definitely part of my todo list ! ;) Which AI voice provider do you use ?

@@Smail-AI vapi

Yeah it definitely depends on the task. My clients have been happy so far with the AI voice bots I delivered to them ;) They knew about the pricing and everything, so I assume they made their ROI and cost/benefit analysis.

Good question ! I did not but it shouldn't be difficult to set up

I think for France Vonage should be used instead of Twilio. If I remember they have French numbers available.

The add-on you're supposed to find in google sheets ? It shouldn't matter whether you're using a Mac or Windows because google sheets is on the browser.

yes with Vapi it is possible because you can use a cloned that you can create in 11labs.

yeah.. you just clone the voice...

Thank you so much !

Thanks for your feedback ! Yes indeed to make the video shorter I just showed how you can host the server in your personal computer. But hey a server can be hosted anywhere, like a VPS for example ;) Would you like to see an upcoming video where I show how to host everything on a remote server ?

Thank you !

The costs involved would depend on the specific AI models chosen (LLM, transcriber, Voice) as well as the number of minutes consumed per month. For this video it costs $0.13/min so if the agent talks for 100 minutes per month then it would be $13/month. Other operating costs are related to the VPS server, around $20/month.

Haha yes I did not spend a lot of time trying to optimise the voice. You can choose other voices that suit you best. I found the best ones are the ones from 11labs and cortana.

Yes this video was made to simulate the process. But in real life you would need a real FB ad campaign with real leads coming in. I might do a future video with real leads being called by an AI voice bot ;)

Thanks for the feedback ! Actually this solution is for Ecom websites that either don't have this information or it is not up to date. Also, the AI receives calls 24/7 so you could call it during the night to have the info if you want. But I agree with you that if the Ecom website already have that info it would be faster to just check that in the product page ;)

You download my resources from the website and then you just follow the tutorial ;) For making the actual call you'll have to press the button "Qualify Seller with AI"

Last time I tried Bland AI was few months ago, are they good now ?

@@Smail-AI same few months ago, still very expensive cuz twillo integration for #s outside US

My pleasure ! Yes there is a way to avoid ngrok, the solution is to use a VPS. That's what I use for real clients. Ngrok is just a tunnel between your local server and the web.

Thank you !

Thanks !

Thanks ! What's your platform ?

Thank you !

The way to recall them automatically will depend on how you get notified about them wanting a refund. If it's by email then you need to setup a webhook that is waiting for incoming mails. If it's your CRMs then you need to connect to your CRM's API etc... Feel free to join the discord server and to ask there I'm pretty sure you'll find your answers there ;)

Hello, derien ! No I'm Moroccan ;)

​@@Smail-AI Excellent, best people for their kindness. Thanks for sharing and don't hesitate to share other skills.

This is quite complex to get right, but a first step would be to create a custom GPT and to feed it this file and to start a test-tune loop feedback until you get something satisfying. More advanced/accurate solutions would require coding knowledge.