Raspberry Rag

C+ 75 completed
Ai Ml
unknown / markdown · small
64
Files
4,374
LOC
0
Frameworks
5
Languages

Pipeline State

completed
Run ID
#372616
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Skipped
Decision
skip_scaffold_dup
Novelty
28.45
Framework unique
Isolation
Last stage change
2026-04-16 18:15:42
Deduplication group #47562
Member of a group with 1 similar repo(s) — canonical #2677 view group →
Top concepts (2)
Project DescriptionData/ML
Open data scored by Repobility · https://repobility.com

AI Prompt

Create a fully offline, open-source speech-to-speech voice assistant designed for university departments, specifically for a Raspberry Pi 5. The system should take voice input, process it through a RAG pipeline, and respond with synthesized speech. The core functionality involves detecting a wake word ("Окей кафедра"), using Vosk for Speech-to-Text (ASR) in Russian, performing search and generation using embeddings (rubert-tiny2) and a vector store (FAISS + SQLite), and finally synthesizing the response using Piper for Text-to-Speech (TTS). The setup should be manageable via a script that handles dependencies, model downloading, and systemd service setup.
python raspberry-pi voice-assistant rag offline speech-to-speech vosk piper faiss linux
Generated by gemma4:latest

Catalog Information

This project is a speech-to-speech voice assistant for university departments on Raspberry Pi 5 (4GB RAM), designed to be fully offline and open-source.

Description

The voice assistant allows users to interact with the system by pressing a button or saying 'Okay department', then asking a question, and receiving an answer in speech. The system uses a knowledge base from documents stored on the Raspberry Pi. It is built using Python, Bash, and NumPy, and does not require any external databases.

الوصف

يعد هذا المشروع مساعدًا صوتيًا للاستفسار عن طريق الصوت، مصمم خصيصًا للقسم الجامعي على Raspberry Pi 5 (4 جيجابايت من ذاكرة الوصول العشوائي)، وتصميمه كاملاً بدون اتصال بالإنترنت ومفتوح المصدر. يتيح للمستخدمين التفاعل مع النظام عن طريق الضغط على زر أو قوله 'أوكي القسم'، ثم سؤاله، ثم الحصول على الإجابة الصوتية. يستخدم النظام قاعدة المعرفة من الوثائق المخزنة على Raspberry Pi. يتم بناؤه باستخدام Python و Bash و NumPy، ولا يحتاج إلى أي قواعد بيانات خارجية.

Novelty

7/10

Tags

speech-to-speech voice-assistant offline open-source university-department raspberry-pi

Technologies

numpy

Claude Models

claude-opus-4.6

Quality Score

C+
74.9/100
Structure
72
Code Quality
100
Documentation
44
Testing
60
Practices
72
Security
92
Dependencies
60

Strengths

  • Good test coverage (31% test-to-source ratio)
  • Consistent naming conventions (snake_case)
  • Good security practices \u2014 no major issues detected
  • Properly licensed project

Weaknesses

  • No CI/CD configuration \u2014 manual testing and deployment

Recommendations

  • Set up CI/CD (GitHub Actions recommended) to automate testing and deployment
  • Add a linter configuration to enforce code style consistency

Security & Health

4.1h
Tech Debt (C)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (2)
Repobility — the code-quality scanner for AI-generated software · https://repobility.com
MIT
License
0.8%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

markdown
63.5%
python
31.1%
shell
2.8%
text
1.6%
yaml
1.1%

Frameworks

None detected

Concepts (2)

Findings produced by Repobility · scan your repo at https://repobility.com/scan/
CategoryNameDescriptionConfidence
Want this analysis on your repo? https://repobility.com/scan/
auto_descriptionProject DescriptionSpeech-to-speech голосовой ассистент для университетской кафедры на Raspberry Pi 5 (4GB RAM). Полностью оффлайн, open-source.80%
auto_categoryData/MLdata-ml70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/96833.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV