ARconvo

Offline Arabic–English bilateral speech translator for bilingual workshops and transcript records. It runs on a PC with no internet connection.

Faster-Whisper
Local translation models
Speech synthesis
Offline

View PRD Discuss this work

Purpose

ARconvo supports two-way Arabic–English spoken communication entirely offline, making it practical for bilingual workshops and transcript records where low connectivity or privacy-sensitive environments rule out cloud inference.

My role

Builder

Highlights

Automatic language detection and silence detection
Faster-Whisper speech recognition with an Arabic-dialect-trained model
Falcon 3B translation with native and model-based speech synthesis
Browser controls, bilingual transcript panels and live system-status visibility
Optimised to run on a PC with 4 GB VRAM

Capabilities

Fully offline local-PC operation
Automatic language and silence detection
Faster-Whisper speech recognition
Arabic-dialect-trained speech workflow
Falcon 3B translation
Windows-native English TTS and model-based Arabic voice synthesis
Bilingual Arabic and English transcript capture
Browser-based controls, live transcript panels and system-status visibility

Governance & quality

Audio and transcripts stay on the PC; no cloud inference
Iterative evaluation with native Arabic-speaking testers
Validation against Arabic detection, translation accuracy and interaction quality

Architecture & delivery

Optimised for a PC with 4 GB VRAM
Designed around local inference and offline edge deployment constraints

Current stage & roadmap

Meeting-platform integration for bilingual workshops and training sessions is planned

Outcomes

Working local application for bilateral offline speech translation
Demonstrates practical voice AI under constrained hardware conditions

Skills demonstrated

Conversational AI
Voice AI
Multilingual NLP
Arabic dialect handling
Speech recognition
Voice synthesis
Translation
Edge AI
Offline inference
Model optimisation