This Free AI Voice Automation sends PDFs Very Fast - Beginner's Guide

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
hey there this is Lea from solar AI I'm calling you because you just opted in our Facebook ad yes well if you're interested in knowing more about the solar potential of your house we can send you a customized analysis for your house by email yeah sure that would be great perfect to do that I just need to know your address and your monthly electricity bill let's start with your address okay first of all we can see that the address is pretty spoton it's exactly the address that we told to the AI as well as the satellite imagery of that hey guys in this video I'm going to show you how you can build an automation that uses an AI voice system and we'll call leads and we'll send them a PDF like the one that you just saw while the call is still happening by the way if you're new to this channel I basically help people make money with AI and automation systems so the tools I'll be using are make.com retail Ai and API template to create the PDFs on the Fly some time ago I watched Liam otley's video where he built a solar bot that gives solar information about a property when I watched it I was thinking okay it's good but there is definitely a way to make it better let video used the Google solar API which is an API that Google released in August 2023 by the way an important detail about the solar API is its coverage so as you can see here those are the areas that are covered so if you want to try to apply this to a property that is in in India this won't work however for all the densely populated areas in the US or in Europe like big cities and so on they have high resolution satellite imagery so it should work very good I think it's a pretty cool API that gives a lot of insight about buildings but first let me show you exactly what will be building imagine a solar agency running Facebook ads a lead will put his name phone and email then this will trigger a phone call from an AI voice bot the Bots will ask whether the lead wants to receive the solar potential of their property and if the answer is yes the AI will ask two questions first it will ask what's the monthly electricity bill and then it will also ask what the specific address of the property based on those two questions it will create a PDF then send it by email to the lead while the call is still happening this PDF contains multiple incentives that the lead May benefit from uh the number of panels that the Google solar API advises to install the payback period as well as the savings during the first year and it will also show a satellite view of the lead's house with a solar arance of the hottest month so basically it's customization instantly and on demand and the goal of this is to really increase conversion rates from Facebook ads this will be a step-by-step tutorial so it should work if you follow exactly the steps that I will be showing you and by the way the me blueprints as well as the code are available in my Resource page so feel free to download it and to test that with your own house the first thing that you want to do is to make sure that retail AI is working on your computer if it's not the case yet well don't worry about that because I created a video where I showed you how you can do just that the next step now is to open the folder where you downloaded your files and then you will right click you will click on opening terminal and you will write pip install requirements 2.txt actually this command will install new python libraries that you need for this video and those are the python libraries that will be installed NPI T file matplot lib pillow and super image the next step now is to test this new library called super image it is basically a library that contains neuron networks that are used to upscale images because if you directly get get the images from the solar API those will be very small so we need a way to upscale them without losing quality so what you will be doing is uh just opening this file test super image. py and then run it I have here this image called image test.png and it will just process it and uh use n network to upscale it now the reason we do that is that the first time you will be running this Library this super image Library it will download the N Network model and it might take some time that's why uh we just make sure that we do this step before so that when we will need it later on to generate the PDF it will not take that much time you know to upscale an image and by the way this Library contains multiple n networks I chose uh this one called Dr Ln because based on some benchmarks it showed the most promising results now I just run this file and uh this is the result you you can see that the image is still the same but it's upscaled with a higher resolution now after that the first thing that we want to do is to create a make scenario because we want to trigger the phone call to the lead so this one will start with the web hook now in the files that you downloaded uh you will just open call solar le. HTML and here you will put the URL of your web hook then you will open the file in your web browser here at this point uh doesn't matter what type of data you will put in because we just want the web hook to know that it will be receiving a name a phone and an email so I just wrote test here and then I trigger the call the result here is that the web hook has now successfully determined the data structure then you will press okay now we need to create a Jon that will serve as the body for our request uh for retail ai's API so this Json will contain three it items the first one is phone number it is required the second item is a name and the third item is an email and require two then you click on save so those three Fields will appear here and then you will select for each field its corresponding data from the web hook and then once it's done you click on okay now the final step is to use the HTTP module the make a request action now before we go any further we need to uh make sure the retel AI server is up up and running if your retail AI server is running you will see your enr address here so you just copy it you paste it here and then you write slash make phone call slash here for the method it will be a post will be needing a header so you just click here add a header for the name uh you will write X retail signature and for the value you will write your uh retell AI API key you can easily find it in your dashboard now for the body type will choose Row the content type will choose Json and for the request content we just uh select the Json string and then we click on okay now we'll test that the call is working so now you just put uh your information of course make sure that your phone number is correct and it starts with a plus one here for the email it doesn't matter whether it's a good or bad email at this point you first run this automation wait for new data so once everything is ready you should click on call the lead now you should be receiving a call there you go now that the call worked we need to move on to the next steps which is to create our PDF template and to do that you need to go to API template. iio and create your account there they have a free tire so you'll be able to create up to three templates and to generate up to uh 50 PDF now you click on manage templates here it should be empty but uh basically you just click on new PDF template and don't worry about all of this guys because this will be available in my Resource page so to create a PDF we need three components the first one is the HTML uh which means the structure of the PDF the second part is the CSS which means The Styling and the colors and the format for this PDF the third component is the Jon part which is the data that will be inputed in that PDF now the next step is to create our second me scenario and this will also start with the web hook because it will be triggered while the call is still happening with the AI now for this web hook we'll be adding it a data structure so you just click on edit here and then show advanced settings and then here we'll click on ADD for the items just follow along and create the same items so those are the items that you need the name the email the phone uh monthly bill Federal incentive State incentive utility incentive uh the number of the panels payback period savings year one address and Ence b64 check them all as required and then you click on Save and then save again and then okay now we need to add the module that will actually create the PDF you just search for API template. and you will click on create a PDF now of course you will need to establish a connection with API template and to do that you need to get the API key from your dashboard and to get that you just go back to uh your templates your PDF templates you close this window you click on this tab here and you will see your API key here now the payload type will be key value pairs the template ID is just the name of your template I called it my new template for the key value pairs we need to add all the data that will be used by API template uh to write the PDF so here you will click on add item all of this data that you see from the web hook will be used in this uh module so for example for the name I will write here name and for the value it will be the actual value of the name I will do the same for the email and you follow the same logic for the rest of the data and you keep the rest of the options as they are which means for the expiration never uh for the delay never as well for those colors CM YK you should uncheck that uh never for the image resample resolution and for the PDF standard you just set it as default and then you click on okay now this module will actually generate a PDF and St it in API template servers but then we need a way to download this file and to do that we need to use the HTP module and you need the action get a file for the URL we'll be using the download URL from the API template module then you click on okay now the final step is to use Gmail you will click on send an email so that you will be able to send a mail from your Gmail account to any type of email actually uh the leads email in my video how to catch all your Facebook leads with AI I showed you how to set up Gmail in make.com so just follow along that video so that you'll be able to do the same and run this tutorial if you have done it already it will be super easy for you to connect Gmail withme.com you just have to click on ADD and then here show advanced settings and uh copy and paste your client's ID and client secret and then once you're done you just click on sign in with Google and choose the right Gmail account that is added as a user for that Google Cloud project now for the recipient you will just select email for the subject I'll just write here your solar analysis for the content I write this uh hello name please find attach the solar analysis of your home now the most important part will be to add an attachment to this email and to do that you just click here here add an attachment by default it will select HTTP get a file from the previous module then you click on okay okay great now we have done our make scenarios now let's test everything actually before we go ahead and start testing this bot we need to do two steps the first one is to copy this web hook URL and to paste it here in uh line 272 of uh this file here now the second step is to get our Solar API key uh from our Google Cloud project to do that you'll just create a new project in your Google Cloud console now that you created the project you just click on apis and services and then you click on enable apis and services here you will just look for two apis the solar API and the geoc coding API both need to be enabled so here's the solar API you click on it and then you click on enable and then you search for Jo coding API you enable it as well now you need to go to credentials here we will create our API key so to do that you click on create credentials and you select API key so once everything is done you'll just have here an API key and you show it uh by clicking here and then you copy it and you paste it here so basically this API key will enable us to use both the solar API to get the insight about buildings but also to use the geocoding API uh which serve to uh translate an address to a latitude and longitude okay so for this particular test we'll be using this address 1643 cner Avenue San Jose California let's first run the two scenarios wait for new data and here let's fill the form with the right data now let's call ourself [Music] uh hello hey there this is Lila from solar AI I'm calling you because you just opted in our Facebook ad yes well if you're interested in knowing more about the solar potential of your house we can send you a customized analysis for your house by email yeah sure that would be great perfect to do that I just need to know your address and your monthly electricity bill let's start with with your address could you please tell me what's the address of your house okay sure it's uh 1643 cner Avenue San Jose California thanks and on average how much do you pay per month on your electricity bill uh on average I pay around [Music] $120 awesome let me send you our analysis by email hold on a sec sure okay now the bot was supposed to confirm that the email has actually been sent I think it just needs a few tweaks on the prompting so now let's check our mail okay first of all we can see that the address is pretty spot on it's exactly the address that we told to the AI as well as the satellite imagery of that if you look at this picture it's exactly the same property and of course we have access to all the information uh so the federal incentives which is about uh two uh $2,500 uh the number of panels advised five the payback period and the savings during the first year so that's awesome it's working great all right guys that was it for this video I hope you enjoyed it and learned something if you have any remarks don't hesitate to comment down below I will be answering you by the way with my AI agency we develop any types of Automation and AI systems if you're interested book a meeting to the currently Link in the description if you like this video make sure to like subscribe and hit the notification Bell it will really help my channel also if you need help for debugging don't hesitate to comment down below cheers guys
Info
Channel: Smail AI
Views: 928
Rating: undefined out of 5
Keywords:
Id: WKhYPia2oXE
Channel Id: undefined
Length: 16min 52sec (1012 seconds)
Published: Sat May 04 2024
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.