Add dedicated MP3 file download component and parallelize large text translation

Fix: improve Tamil female TTS voice quality using premium customization

Feat: expose processing mode inside Text to Speech tab

Feat: integrate Meta MMS Tamil TTS with young female voice pitch-shifter

Refactor: clean up unused chat engine, easyocr fallback, and character memory UI

Enhance UI with premium dark-mode styling, HTML header, and force-dark JS listener

Upgrade Cloud Translation to Qwen-72B via HF Inference API for high-quality conversational Tamil

Implement Custom Translation Dictionary Editor UI and JSON persistence

Add common English loanwords (Story, Happy, Try, Help, Thanks, etc.) to colloquial mapping for natural spoken Tamil

Support Spicy Mode (Intimate Translation) checkbox in the Text to Speech tab

Optimize BGM downloading with parallel ThreadPoolExecutor and reduce output MP3 bitrate to 64k mono for 3x faster client downloads

Support modern Tamil translations with comprehensive case suffixes, pronouns, and verb endings

Fix event loop already running error in fallback TTS, improving stability and speed inside Gradio

Scale TTS to 50k+ characters, add recursive parallel chunking, resolve openvino IDE warning

Optimize voice pipeline: add fast Cloud APIs mode, lazy load models, and run concurrent panel tasks

Fix import logic to catch AttributeError and other exceptions, preventing startup crash on HF Space

Update translation engine to follow 5-stage hybrid workflow (Lingua + IndicTrans2 + Qwen3)

Refactor and optimize Tamil voice pipeline with expressive style fallback

feat: implement robust multilingual translation voice pipeline with glossary, memory database, and unified Gradio translator UI

Optimize MP3 compression and parallelize BGM downloads for 4x faster audio downloads

Migrate to local Qwen3, IndicTrans2, and Svara-TTS OpenVINO INT4 pipeline with robust fallbacks and lust style

Update Dockerfile to run lightweight Gradio application

Add lightweight Dockerfile for Hugging Face Spaces deployment

Remove Video Dubbing Studio to make app lighter and smoother

Optimize audio generation, serve local BGM, deliver compressed MP3 files

Change Vite base to relative ./ for robust iframe asset loading

Add Safari-like Reader Mode and highlight-to-populate selection logic to Text to Speech tab

Add Import Text from URL feature with line-based and phrase-based cropping options

Fix frontend build copy path in Dockerfile

Handle invalid/empty start_page_num inputs in jump_to_page function safely

Improve auto-play ended event listener with addEventListener and element property check

Support custom start and end page ranges for comic reading mode

Support proper automatic switching and audio autoplay in comic reader mode

Fix setUploading and uploading ReferenceErrors in App.jsx

Configure Vite build outDir and base path for FastAPI, add keep-alive scripts

feat: add heavy GPU dependencies and Wav2Lip auto-downloader for Colab deployment

feat: unify React Dubbing Studio UI with Gradio Manga AI in single Docker container

remove: delete 'Soft & Seductive (Maya)' voice tone from all dropdowns

fix: replace localhost iframe in dubbing tab with native Gradio UI that works on HF Spaces

revert: switch HF Space back to Gradio SDK running app.py (Maya Immersive Tamil Manga AI)

fix: remove root GET route blocking React frontend on HF Spaces

fix: upgrade Node.js 18 to 20 to satisfy Vite 8 build requirements

fix: switch HF Space SDK from gradio to docker to use unified Dockerfile

fix: ensure static files are mounted last to prevent API route interception

fix: unified Docker deployment for Hugging Face to resolve local network connection errors

feat: further refine Tanglish vocabulary and common spoken verb forms

feat: enhance translation with colloquial (spoken) Tamil style

feat: implement detailed pipeline diagnostics, memory optimization for transcription, and fix import issues

Implement Cinematic Panel Narration with TTS splicing

Implement 2-stage PaddleOCR + Mistral pipeline