Offline Conversational LLM on Pi

$750-1500 USD

Posted 1 minute ago

•

Ends in 6 days

$750-1500 USD

Paid on delivery

I’m turning a plush toy into a bilingual chatterbox that never needs Wi-Fi. Inside the toy sits a Raspberry Pi 5 with 8 GB of RAM; your task is to pull an open-source large-language model (roughly 8 GB–24 GB) from Hugging Face, streamline it, and make it run locally on that board. The goal is simple spoken conversations in both English and Spanish, so the full loop—speech-to-text, prompt handling, text-to-speech—has to live on the Pi. I already have a fresh 64-bit Raspberry Pi OS install, a USB microphone/speaker HAT, and plenty of storage. What I need from you is: the model selection, any quantisation or pruning that keeps RAM usage within limits, all required runtime libraries ([login to view URL] / GGUF, [login to view URL], or whatever toolchain you prefer), and clean startup scripts that boot straight into conversation mode when the toy is powered on. Latency should feel natural enough that a child can ask a question and hear a response without noticeable lag. Deliverables • The optimised model files or exact download links • Installation and build scripts with inline comments • A single launch script or systemd service that autostarts the conversation loop • Brief README explaining how to switch between English and Spanish during use I’ll flash the image you provide, power the toy, and run a quick two-minute demo in both languages; if it holds the conversation without network access, the job is done.

Project ID: 40816526

About the project

35 proposals

Open for bidding

Remote project

Active 5 secs ago

Place your bid

Bid amount Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

It's free to sign up and bid on jobs

35 freelancers are bidding on average $1,002 USD for this job

User Avatar
Flag of UKRAINE

Hello, My Raspberry Pi and embedded-systems experience makes me well suited to turning your plush toy into a self-contained conversational device. I understand the practical constraints of a Pi-based product, including local resource usage, USB audio integration, startup behaviour, and reliable operation without network access. I can work with the Raspberry Pi 5 environment and focus on an offline bilingual conversation loop using suitable open-source components for speech recognition, local model inference, and text-to-speech. My background in electronics, embedded systems, automation, and complete product development is valuable here because the result must work as an integrated physical device, not merely as an isolated model. I will provide the optimised model files or precise download instructions, commented installation/build scripts, an autostart launch script or systemd service, and a concise README covering English/Spanish switching. I would be glad to build and validate this offline Raspberry Pi implementation for your two-minute demo. Thanks!

User Avatar
Flag of SERBIA

HELLO, "Offline Conversational LLM on Pi" - you need a genuinely offline English-Spanish voice assistant running smoothly on a Raspberry Pi 5 with only 8 GB RAM. I'll handle the complete pipeline: selecting a suitable lightweight open-source model, quantising it for the Pi, configuring local speech recognition and bilingual text-to-speech, and connecting everything into a responsive conversation loop. I'll also optimise memory usage and inference latency, since running an LLM alongside speech processing on 8 GB RAM is the main constraint. I'll provide commented installation scripts, model links, an automatic startup service, and a README with language-switching instructions. Should English and Spanish be selected automatically from the speaker's language, or switched using a voice command? I appreciate your consideration and look forward to making your toy work entirely offline. THANKS

User Avatar
Flag of INDIA

Hello Sir, I can efficiently transform your plush toy into a bilingual chatterbox using a Raspberry Pi. I will select an appropriate open-source LLM, optimize it with quantization, and ensure it runs smoothly for speech-to-text and text-to-speech in both English and Spanish. My approach includes setting up the necessary libraries and creating startup scripts for seamless operation. I am ready to start right away. What specific features would you like to prioritize during implementation? Thanks, Deepak

User Avatar
Flag of UKRAINE

Hi! I’d start by reviewing your Pi setup and then map the offline speech loop for English and Spanish. With your Raspberry Pi 5, I can select a lean open-source model, streamline it, and wire speech-to-text, prompt handling, and text-to-speech to run locally. My Python and Machine Learning (ML) experience fits the model optimization and low-latency runtime work here. I’ll reduce RAM use, package the build scripts, and set up a clean startup flow so the toy boots straight into conversation mode. I’ll handle planning, testing, and handover, with friendly support for fixes, updates, and maintenance after delivery.

User Avatar
Flag of AUSTRALIA

Hello, This is exactly the kind of embedded AI project I enjoy working on. I have hands-on experience with **C programming, Python, and Machine Learning**, and I’ve built similar offline voice-driven systems where everything had to run locally without relying on the cloud. For your Raspberry Pi 5 plush toy assistant, I can help you choose a model that’s actually practical for the hardware, then optimize it using **GGUF / quantization**, and wire the full pipeline together with **speech-to-text, prompt handling, and text-to-speech** so it boots straight into conversation mode. I’ve worked on a recent similar setup where I integrated a lightweight local LLM with Whisper-based speech recognition and a TTS loop on a Raspberry Pi for an offline assistant prototype, so I’m familiar with the memory and latency constraints you’re dealing with. I’d also make the English/Spanish switching simple and child-friendly, so the toy feels natural to use. One thing I’d want to confirm: do you prefer a fully voice-detected language switch, or a simple spoken command like “English” / “Spanish”? If you’d like, I can walk you through the best model choice and startup flow on a quick call. Best regards, Mehreen

User Avatar
Flag of UKRAINE

Hello, My Raspberry Pi and embedded-systems experience directly matches your offline conversational toy, including the need for reliable boot-time operation on a Raspberry Pi 5. I can work within the board’s 8 GB RAM constraint and deliver a self-contained bilingual conversation loop rather than relying on cloud services or Wi‑Fi. My background in Embedded C/C++, firmware, driver development, RTOS, Raspberry Pi, and audio-capable product development is well suited to integrating the microphone/speaker HAT with local speech recognition, model inference, and speech output. I can also provide the required installation/build scripts, commented startup service, model files or precise download sources, and a concise README covering English/Spanish switching. I will focus on a practical, maintainable deployment that starts directly into conversation mode and remains fully offline. Choose me to turn the Raspberry Pi hardware into a dependable bilingual chatterbox. Thanks!

User Avatar
Flag of UNITED STATES

Hello, My Arduino experience is directly relevant to creating reliable embedded behavior for a physical interactive toy, including startup handling and communication with connected hardware. I can help build the Raspberry Pi 5 conversation system around that practical device-focused foundation, with attention to offline operation and a simple child-friendly interaction. I can configure the required local runtime and startup behavior for the Pi, connect the USB microphone and speaker HAT, and provide commented installation scripts, model download references, a launch service, and a README covering English and Spanish switching. The finished setup will be designed to operate without Wi-Fi and to make the boot-to-conversation experience straightforward. I would be glad to apply my embedded development skills to turn your plush toy into a self-contained bilingual chatterbox. Thanks!

User Avatar
Flag of UNITED STATES

Hello, Building an offline conversational system on Raspberry Pi requires strong Linux and low-level performance skills, and my background in Python, C programming, and Linux directly matches that need. I can help turn the Pi 5 into a self-contained bilingual voice assistant using an appropriate local model and lightweight runtime. My AI engineering experience includes LLM integration, intelligent assistants, agent orchestration, and production-ready deployment. Combined with C and Linux expertise, this supports the practical parts of your project: selecting a model that fits the Pi’s memory, using quantised local inference, integrating offline speech-to-text and text-to-speech, and creating installation scripts, startup automation, and clear language-switching instructions. I will deliver the optimised model references or files, commented build and installation scripts, an autostart launch script or systemd service, and a concise README for the English/Spanish conversation flow. I am ready to build this into a reliable Wi-Fi-independent prototype for your plush toy. Thanks!

内容来自 Freelancer News,版权归原作者所有。

阅读原文 ↗

把喜欢的内容,留在你的新标签页。

打开我的订阅