prikaz prve stranice dokumenta GPT virtualni assistent implementiran na ESP32 platformi
  Download
PDF 732.9 KB
undergraduate thesis
GPT virtualni assistent implementiran na ESP32 platformi
Rijeka: University of Rijeka, Faculty of Engineering, 2023. urn:nbn:hr:190:749701
Šćulac, Robin
University of Rijeka
Faculty of Engineering
Department of Computer Engineering
Section of Intelligent Computing Systems

Institutional repository: Repository of the University of Rijeka, Faculty of Engineering

Cite this document

Šćulac, R. (2023). GPT virtualni assistent implementiran na ESP32 platformi (Undergraduate thesis). Rijeka: University of Rijeka, Faculty of Engineering. Retrieved from https://urn.nsk.hr/urn:nbn:hr:190:749701

Šćulac, Robin. "GPT virtualni assistent implementiran na ESP32 platformi." Undergraduate thesis, University of Rijeka, Faculty of Engineering, 2023. https://urn.nsk.hr/urn:nbn:hr:190:749701

Šćulac, Robin. "GPT virtualni assistent implementiran na ESP32 platformi." Undergraduate thesis, University of Rijeka, Faculty of Engineering, 2023. https://urn.nsk.hr/urn:nbn:hr:190:749701

Šćulac, R. (2023). 'GPT virtualni assistent implementiran na ESP32 platformi', Undergraduate thesis, University of Rijeka, Faculty of Engineering, accessed 01 June 2024, https://urn.nsk.hr/urn:nbn:hr:190:749701

Šćulac R. GPT virtualni assistent implementiran na ESP32 platformi [Undergraduate thesis]. Rijeka: University of Rijeka, Faculty of Engineering; 2023 [cited 2024 June 01] Available at: https://urn.nsk.hr/urn:nbn:hr:190:749701

R. Šćulac, "GPT virtualni assistent implementiran na ESP32 platformi", Undergraduate thesis, University of Rijeka, Faculty of Engineering, Rijeka, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:190:749701

Metadata
TitleGPT virtualni assistent implementiran na ESP32 platformi
Title (english)GPT Virtual Assistant Based on ESP32 Platform
AuthorRobin Šćulac
MentorMladen Tomić (mentor)
Committee memberMiroslav Joler (predsjednik povjerenstva)
Committee memberMladen Tomić (član povjerenstva)
GranterUniversity of Rijeka
Faculty of Engineering
(Department of Computer Engineering)
(Section of Intelligent Computing Systems)
Rijeka
Defense date and country2023-09-19, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Architecture of Computer Systems
Abstract
Ovaj rad istražuje simbiozu ugradbenih sustava i napredne umjetne inteligencije pomoću čipa ESP32-C3 za stvaranje inovativnog virtualnog pomoćnika. Koristi Bare Metal Rust za učinkovitu kontrolu hardvera i uključuje Whisper i LLAMA 2 AI za visokokvalitetnu konverziju govora u tekst i razumijevanje prirodnog jezika. Bežične značajke i značajke niske potrošnje energije ESP32-C3 čine ga idealnim. Bare Metal Rust optimizira performanse, memoriju i vrijeme odziva, što je ključno u postavkama s ograničenim resursima. Studija uspoređuje Rust s etabliranim jezicima, ocjenjujući njihove značajke i sigurnost memorije. Komunikacija se oslanja na GATT za razmjenu podataka, poboljšavajući prepoznavanje govora putem Whispera. LLAMA 2 dodaje svijest o kontekstu, poboljšavajući prepoznavanje namjere korisnika i relevantnost odgovora. Dvosmjerni protok podataka poboljšava interakciju. Korisnički unos neprimjetno dolazi do poslužitelja za informirane odgovore. Dokument detaljno opisuje implementaciju, pokrivajući hardver, Rust programiranje, postavljanje GATT-a, integraciju Whispera, uključivanje LLAMA 2 i uspostavljanje dvosmjerne komunikacije. Procjena učinka naglašava učinkovitost, osjetljivost, preciznost i korištenje resursa. Koristeći ESP32-C3 i Rust, ovaj rad prikazuje integrirani pristup i transformativni potencijal kombiniranja hardverske ekspertize s umjetnom inteligencijom.
Abstract (english)
This work explores the fusion of embedded systems and advanced AI using the ESP32-C3 chip to create an innovative virtual assistant. It employs Bare Metal Rust for efficient hardware control and incorporates Whisper and LLAMA 2 AI engines for high-quality speech-to-text conversion and natural language understanding. The ESP32-C3's wireless and low-power features make it ideal. Bare Metal Rust optimizes performance, memory, and response time, crucial in resource-constrained setups. The study compares Rust with established languages, evaluating their features and memory security. Communication relies on GATT for data exchange, enhancing speech recognition via Whisper. LLAMA 2 adds context-awareness, improving user intent recognition and response relevance. The two-way data flow enhances interaction. User input seamlessly reaches the server for informed responses. The paper details implementation, covering hardware, Rust programming, GATT setup, Whisper integration, LLAMA 2 inclusion, and establishing bidirectional communication. Performance evaluation underscores efficiency, sensitivity, precision, and resource use. Using ESP32-C3 and Rust, this work showcases an integrated approach and the transformative potential of combining hardware expertise with AI.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:190:749701
Study programmeTitle: Computer Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence Attribution 4.0 International (CC BY 4.0) icon
Created on2023-09-26 14:31:43