Version 1.5.0-beta01

This release include Real time Transcription feature, allowing users to transcribe their recordings in real-time (#23). It includes a comprehensive system for managing speech-to-text (STT) models, background downloading, and a refined UI for displaying live results.

✨ New feature

• Live Transcriptions ((Requested in #23)): Real-time speech-to-text conversion during recording sessions.
• On-Demand Model Downloads: Keeping the app size into account , no models are pre-installed with the apk
• Model Management: A new dedicated transcription settings tab allows users to download, delete, and switch between different language models (using Vosk).
• Controlled Visibility: Transcription results can be toggled on/off on the go, even during an active recording.
• Bookmarks with Transcription: Ability to add bookmarks directly from the recording screen with integrated transcription info.

🛠️ UX Improvements

• Live Feedback: Added a notification system to show the progress of model downloads.

🛠️ Internal & Infrastructure

• AGP Upgrade: Upgraded Android Gradle Plugin to version 9.0.
• Improved Concurrency: Refactored TranscriptionProviderImpl to use non-blocking buffers and more responsive data flow.
• Robust Storage: Integrated Okio for efficient file handling and unzipping of model files.
• DI Refinement: Optimized dependency injection scopes for recorder service and transcription providers.

📦 Dependencies

• Updated project dependencies and library versions.
• Ktor client for model downloads.
• Integrated Vosk for offline speech recognition.

🚀 Future-Proofing & Data Persistence
In addition to these updates, looking forward with adding post-transcription workflow. Currently the transcription results are not saved in the system later this will be added.

Full Changelog: v1.4.4...v1.5.0-beta01