SIRENE — AI-Powered Filipino Voice Contribution & Language Learning Platform
SIRENE is an AI-powered multilingual language learning platform that helps users learn Philippine languages through interactive translation exercises, AI speech evaluation, and gamified vocabulary challenges while generating valuable AI-annotated speech datasets for Philippine language research.
An illustrative 3D model made for this portfolio.
Overview
SIRENE addresses two interconnected challenges: making Philippine language learning more engaging and accessible, and increasing the availability of high-quality multilingual speech datasets for underrepresented Philippine languages. The platform allows users to practice Philippine languages through AI-generated translation exercises, vocabulary challenges, and speech activities. Users can record their spoken translations, which are analyzed by AI for pronunciation, fluency, completeness, and translation accuracy. What makes SIRENE different is that learning activities also contribute to the creation of an AI-annotated multilingual speech dataset. Each completed activity can generate structured data such as the source phrase, target language, spoken translation, audio recording, transcription, AI evaluation scores, and speaker metadata. This dataset can potentially support future research and development in Automatic Speech Recognition (ASR), Machine Translation, Speech-to-Speech Translation, Pronunciation Assessment, Speaking Proficiency Evaluation, and Code-Switching Research. SIRENE therefore combines language education, gamification, AI-powered speech analysis, and dataset generation into a single platform.
Stack
Key features
- 🤖 AI-Generated Translation Exercises — Generates interactive translation activities for Philippine language learning.
- 🎤 AI Speech Evaluation — Evaluates pronunciation, fluency, completeness, and translation accuracy.
- 📚 Vocab Blitz — Daily vocabulary challenges for building and reinforcing language knowledge.
- 🏆 Achievements & Progress Tracking — Tracks learning progress and rewards users through gamified achievements.
- 🌏 Multilingual Language Support — Supports learning and practicing Philippine languages.
- 🎙️ Speech Recording & Transcription — Captures spoken responses and generates transcriptions.
- 📊 AI-Annotated Dataset Collection — Converts learning activities into structured multilingual speech data.
- 🎮 Gamified Learning Experience — Uses challenges, achievements, and progress tracking to encourage continued practice.
- 🔬 Research-Oriented Dataset Generation — Produces data that can support future Philippine language AI research.