Live Chat

B 81 completed
Cli Tool
cli / markdown · tiny
44
Files
3,553
LOC
1
Frameworks
4
Languages

Pipeline State

completed
Run ID
#353372
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Skipped
Decision
skip_scaffold_dup
Novelty
37.58
Framework unique
Isolation
Last stage change
2026-04-16 18:15:42
Deduplication group #47666
Member of a group with 1 similar repo(s) — canonical #78652 view group →
Top concepts (2)
Project DescriptionTesting
Repobility — the code-quality scanner for AI-generated software · https://repobility.com

AI Prompt

Create a terminal-based, voice-first conversational AI agent. I need it to listen through the microphone, use VAD (Silero) for speech detection, transcribe audio using Whisper STT locally on Apple Silicon, process the input with Claude, and then speak the response back using local TTS (Piper). The system should support both an API backend (using an environment variable for the key) and a CLI backend. It should also allow configuration via a YAML file to set parameters like fast/deep models, persona, and silence detection thresholds.
python cli ai voice-assistant llm speech-recognition tts terminal claude apple-silicon
Generated by gemma4:latest

Catalog Information

A terminal-based voice-first conversational AI that listens, transcribes, processes with Claude, and speaks back using local TTS.

Description

This program functions as a voice-first conversational AI that runs entirely from the terminal. It begins by listening through the microphone, detecting speech boundaries with VAD, and transcribing the audio locally using Whisper. The transcribed text is sent to Claude, which generates a response streamed token-by-token. The response is then converted to speech locally via Piper TTS and played back through the speaker. Users can configure model choices, TTS voice, and other settings through a simple YAML file, and the tool supports both API and CLI backends for Claude.

الوصف

يعمل هذا البرنامج كمساعد ذكاء اصطناعي يعتمد على الصوت ويشغل بالكامل من الطرفية. يبدأ بالاستماع عبر الميكروفون، ثم يكتشف بداية ونهاية الكلام باستخدام تقنية VAD. يُحوَّل الكلام إلى نص محليًا باستخدام نموذج Whisper، ثم يُرسل النص إلى نموذج Claude لتوليد الرد. يُعاد تحويل الرد إلى كلام بصوت محلي عبر نموذج Piper، ويُعرض الصوت مباشرة على مكبر الصوت. يتيح البرنامج اختيار نماذج مختلفة للردود السريعة أو المعقدة، مع إمكانية التبديل بين واجهة برمجة التطبيقات أو واجهة سطر الأوامر للـ Claude. يمكن تكوين إعدادات مثل صوت TTS، سرعة الاستجابة، ونوع النموذج عبر ملف YAML بسيط. يُعد مثالًا عمليًا لتجربة الذكاء الاصطناعي الصوتي في بيئة سطر الأوامر، مع دعم للمحادثات الثنائية عبر الأجهزة الصوتية.

Novelty

7/10

Tags

voice-interface conversational-ai terminal-assistant real-time-audio local-tts speech-recognition ai-chatbot

Technologies

anthropic numpy pytorch rich

Claude Models

claude-opus-4.6 claude-sonnet-4.6

Quality Score

B
81.1/100
Structure
80
Code Quality
100
Documentation
66
Testing
70
Practices
65
Security
100
Dependencies
60

Strengths

  • Good test coverage (83% test-to-source ratio)
  • Code linting configured (ruff (possible))
  • Consistent naming conventions (snake_case)
  • Low average code complexity \u2014 well-structured code
  • Good security practices \u2014 no major issues detected

Weaknesses

  • No LICENSE file \u2014 legal ambiguity for contributors
  • No CI/CD configuration \u2014 manual testing and deployment

Recommendations

  • Set up CI/CD (GitHub Actions recommended) to automate testing and deployment
  • Add a LICENSE file (MIT recommended for open source)

Security & Health

4.1h
Tech Debt (C)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (3)
About: code-quality intelligence by Repobility · https://repobility.com
Unknown
License
3.1%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

markdown
61.3%
python
37.5%
toml
0.7%
yaml
0.4%

Frameworks

pytest

Concepts (2)

Repobility (https://repobility.com) — every score reproducible
CategoryNameDescriptionConfidence
Repobility · code-quality intelligence platform · https://repobility.com
auto_descriptionProject DescriptionVoice-first conversational AI agent. Listens through your mic, thinks with Claude, and speaks back using local TTS. Runs entirely from the terminal.80%
auto_categoryTestingtesting70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/77485.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV