Back to Projects
fullstackFeatured

Voxia - Real-Time Speech Translation

iOS app combining real-time speech transcription, instant translation across 100+ languages, and AI-powered conversation analysis.

Founder & Full Stack Developer
2025-Present

Screenshots

Voxia - Real-Time Speech Translation screenshot 1

About This Project

Voxia is an iOS application that eliminates language barriers by combining live speech transcription, real-time translation, and AI-driven conversation analysis into a single platform. It supports 100+ languages with features like Interpreter Mode for face-to-face translations, Live Pair for connecting two users with instant translation, and Group/Classroom Mode for shared live captions. The app also includes AI-powered meeting analysis with summaries, sentiment detection, and automatic action item identification.

Key Features

  • Live word-by-word transcription with automatic speaker recognition
  • Real-time sentence-level translation across 100+ languages
  • Interpreter Mode for face-to-face translations between two people
  • Live Pair connecting two users with instant translation
  • Group & Classroom Mode for shared live captions at events
  • AI conversation analysis with summaries and sentiment detection
  • Vocabulary Builder with flashcards and progress tracking
  • Audio/video file import and processing with speaker diarization

Challenges & Solutions

  • Achieving low-latency real-time transcription and translation
  • Supporting 100+ languages with high accuracy
  • Implementing speaker diarization for multi-person conversations
  • Building a seamless interpreter mode for face-to-face interactions

Results & Impact

  • Comprehensive speech translation platform ready for launch
  • Support for 100+ languages with real-time processing
  • Multiple interaction modes for diverse use cases

Technologies Used

Next.jsReactTypeScriptSwiftiOS SDKAI/MLSpeech RecognitionTailwind CSS

Project Details

Category
fullstack
My Role
Founder & Full Stack Developer
Duration
Ongoing
Year
2025-Present
Status
Public