Changelog

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Added

  • Added authenticated /api/v1 routes for user, profile, analytics, transcriptions, folders, templates, uploads, sharing, logout, audio segments, and participant search.
  • Added UUID support for templates, folders, requests/results, and tasks, including migrations that backfill existing records.
  • Added UUID-first API payloads for transcriptions, folders, templates, users, share settings, results, and tasks.
  • Added filtering, sorting, includes, pagination, and sparse fieldsets for key API list endpoints, including only and with aliases.
  • Added richer API documentation examples generated from the current user's data, including request examples for filters, sparse fields, uploads, sharing, and profile updates.

Changed

  • API v1 endpoints now prefer UUID identifiers such as folder_uuid, template_uuid, default_template_uuid, and shared_result_uuids, while keeping numeric IDs available for legacy API compatibility where supported.
  • Upload and sharing endpoints now accept UUIDs for templates, folders, transcriptions, and shared result selections.
  • Transcription API responses now expose structured result metadata, normalized model names, task UUIDs, folder UUIDs, and reduced detail payloads for sensitive settings.
  • Template selection for shared-team transcriptions now loads accessible personal, team-global, system-global, and public team templates more consistently.

Security

  • API v1 responses omit internal numeric IDs and sensitive settings from public mobile/API payloads unless the legacy API surface is used.
  • Share settings validation now verifies selected result UUIDs or IDs belong to the target transcription before saving.

Fixed

  • Transcript pages without an improved transcript no longer request the alignment map endpoint.
  • Alignment map polling now uses a single in-flight request with managed retries instead of stacking repeated browser requests.
  • Alignment map generation is now queued once per transcription while a map is missing, avoiding duplicate backend jobs during polling.

1.12.0 - 2025-01-14

This release represents approximately one year of development since the initial v1.0.0 release, including major new features for team collaboration, speaker management, and visibility controls.

Added

  • User Groups & Visibility System: Complete team sharing infrastructure with granular access controls

    • Create user groups to organize team members
    • Set visibility on transcriptions, folders, and templates (private, team, specific groups)
    • Resource-level access control with inheritance from folders
    • Visibility badges showing sharing status throughout the UI

  • Speaker Management: Post-transcription speaker editing capabilities

    • Rename speakers after diarization is complete
    • Reassign transcript segments to different speakers when detection is incorrect
    • Real-time speaker badge updates in transcript viewer
    • Speakers panel with word count percentages per speaker

  • Folder System Enhancements

    • Folder visibility controls (private, team, shared with groups)
    • Nested folder support with parent-child relationships
    • Drag and drop transcriptions between folders
    • Folder-level default templates

  • Transcript Actions: Unified button component for transcript tab

    • Combined "Improve Transcript" and "Copy Transcript" buttons
    • Consistent styling and positioning

  • Dashboard Improvements

    • Performance optimizations for statistics loading
    • Transcription links in task cards
    • Improved recent activity display

  • Mobile Responsive Updates

    • Speakers panel repositioned above transcript on mobile/tablet
    • Improved layout adaptations across screen sizes

Changed

  • Template visibility now supports private, team, and system-wide options
  • Onboarding flow updated to collect user preferences (industry, meeting types)
  • Share modal auto-saves first version selection
  • Subscription page restricted to team admins only

Fixed

  • Memory exhaustion with large transcripts in speakers panel
  • Shared link access bypassing team scope correctly
  • Folder parent relationship resolution
  • Upload modal state management
  • Transcription visibility in shared folders
  • Various test suite stability improvements

Security

  • Subscription management restricted to authorized team members
  • Improved authorization checks on bulk actions (move, delete)

1.0.1 - 2024-11-XX

Fixed

  • Initial bug fixes after launch

1.0.0 - 2024-11-XX

Added

  • Initial public release with semantic versioning
  • Core transcription functionality with Whisper AI
  • GPT/Claude/Gemini integration for analysis
  • Team management and collaboration features
  • Subscription management with Stripe
  • Template system for structured analysis
  • Real-time processing with Laravel Horizon
  • Secure password-protected sharing
  • Mobile app support with React Native/Expo

Features

  • Audio/video transcription pipeline
  • Multi-model AI processing (OpenAI, Anthropic, Google, Enterprise)
  • Team-based workspace management
  • Customizable analysis templates
  • Export to PDF/DOCX formats
  • Real-time status updates
  • Diarization support for speaker identification