ESP32 Voice Assistant with ChatGPT: Your Personal AI Companion

so finally here comes our most targeted project of making our own voice assistant based on chat DVD let me show you a quick demo of the project please ask who is Dr apj Abdul Kalam Dr apj Abdul Kalam was an Indian scientist and Statesman who served as the 11th President of India from 2002 to 2007. so that was our esp32 chat gbd voice assistant and in this video I'll be guiding you how I make it and how you can make it too but before that you need to watch out the other episodes of the series yes so in this esp32 chat Deputy series we already uploaded three episodes whose link you can find in the description of this video and you need to watch out all three episodes otherwise you won't be able to understand this video and you won't be able to make this project as well so do watch the doubt and if you already watched it well let's get started with making our own voice assistant based on creativity this video is sponsored by LTM and they came up with an amazing product called as ldm 365. so ldm 365 is an Electronics product design platform that unites PCB design mcad data management and teamwork so with LTM 365 you can do the PCB designing task you can share your projects over web for review purposes it do cover sharing your PCB file to Mechanical teams so that they can create the mechanical product package based on your PCB then it also provides the centralized cloud storage so you don't need to rely on one single computer for your files it also helps you with managing your components and get real-time supply chain data for your components it also allows multiple people to work on single project and in the end it also helps you with sending your design to final manufacturing units so Altium 365 takes care of all other tasks so you put more time and effort in making something creative and useful and the good part is you can try out it free version as well I'll leave its free trial version link down in the description of this video so do check that out and now let's start with this video so in the last video we were successfully able to convert our speech to text using Google cloud services and in this video we'll be continuing from that point only so with the last code we were getting the responses in Json format and along with a required answer we were also getting some other responses so we first put the code to filter out the specific text converted from speech after writing the filtering code and uploading them now we are getting only the text of our speech great so now I just need to provide this text question to chat GPD via its API and for that I have copied this code from the first episode of the series and integrated that code in this after modifying it a bit now we can ask questions directly through mic and we can get the answers from chat GPT straight on the serial monitor but we want to listen the answers from the speaker instead of Serial monitor well I know we already covered this part in the second episode of this series but here is one catch so in both the code like this one which converts speech to text and this one which converts text to speech uses a library with the same name called as audio dot at but has totally different internal coding so when we try to merge both the codes we've got a lot of compiling errors so now to solve this we came up with a solution of using two separate ESP board one for speech to text and other for text to speech now let me show you the proper explanation about how this both ESP will communicate with each other for our project on a whiteboard okay so here are both the esps one for speech through text and other for text to speech so in speech to text we have a microphone and an esp32 as a hardware okay and in text-to-speech we have esp32 and a speaker as a hardware now I'll show you what kind of uh like tasks are performed by this section and what kind of tasks are performed by this section so here first of all it will take the input from microphone that's input from mic okay after process after taking the input it will you know give to speech to text service by Google Cloud so it will be requesting that API then we'll be getting the response and that response will be given to chat GPT API chat GPT API and after requesting the question to chat GPT API we'll be getting the answer from chat GPT and will be sending that a response to you art okay so both the esps are communicating why uart communication so in the end we'll be sending the response sending the answer coming from the chat Deputy why you are two other esp32 now for those who don't know you want it's basically the serial communication the rxtx communication okay so these four tasks are performed at this ESP and once we get the response the task of this ESP comes into the picture Okay so first of all it will take the input from uart so whatever response it is sending it will be taken by this after that it will be sending to text to speech conversion okay so we have a built-in function for that so it will go into text to speech conversion after that uh the the response will be given to the speaker as an output okay so these three tasks will be performed by this esp32 and that's how we will be able to make our own voice assistant based on chat with the help of two microcontrollers so here is the final code which we wrote based upon the explanation now let's upload the code in respective ESP boards after uploading the code we made the connections of all the components according to this connection diagram now here we also used an IES sensor that will be acting as a wake up button an RGB LED that will show the status of our project and we also connected a tp4056 battery charging module and a battery to make it work remotely as well then we shouldered all the components on a general purpose PCB and tried to fit it inside a plastic casing we also did two holes on the top for the mic and IR sensor and after complete assembly the vinyl project looks like this now let's start using it okay so to test it out first we'll turn on the switch starting perfect it says starting in now or we are good to go to ask the questions so first I'll wave my hand over the proximity sensor how are you so when the green light turns on it means uh it is ready to on it is it is recording okay when the blue light turns on it means it's sending the data for speech to text and when the red light is on it is send the data to chat gbt API I'm doing well thank you perfect so let us ask another question name is us who is Elon Musk blue means uh speech to text red do means a two chat GPD API and now let's just wait for the response the Elon Musk is an entrepreneur when innovator best known for founding Space X Tesla and PayPal he is also the CEO of both Tesla perfect so we have provided a way less token around like 30 40 tokens so we got the short answer but also thread asking question in Hindi and it responds in Hindi as well let us try to ask one more question foreign the like button right now if you love this video and got to learn something new from this complete series and yes this was kind of a final episode of this series but do let me know your suggestion about how we can do more improvements in our own made Voice Assistant do you have any suggestions like adding something in it well do drop your suggestion down in the comments of the video and we may get a new episode in this series we never know so yeah do subscribe our channel to see more such amazing projects coming up on our Channel and that being said I am just ending this video here and now just wait for my next video Until Then Explorer learn share with me techy SMS foreign

Channel: techiesms

Views: 50,697

Keywords: techiesms, electronics, electronics projects, esp32 chatgpt, chatgpt on esp32, chatgpt voice assistant, chatgpt api, speech to text, text to speech, esp32 projects, iot projects

Id: IvAAw93e-9E

Length: 8min 25sec (505 seconds)

Published: Sat Jul 22 2023