Changelog
Follow up on the latest improvements and updates.
RSS
new
Release
New Features & Major Enhancements
- Expanded Next-Generation TTS Models:Added Fish Audio S2.1 Pro and S2 Pro, MiniMax Speech 2.8 HD and Speech 2.8 Turbo, plus Cartesia Sonic 3.5 and Sonic 3.6. Existing models remain available, while the newest options now appear first for easier selection.
- Expanded Celebrity Voice Library:Added new celebrity AI voices with dedicated voice pages, preview audio, voice characteristics, background information, suggested use cases, and direct access to voice generation.
- Broader Multilingual Voice Generation:Expanded language coverage across the latest Cartesia models, with Sonic 3.6 supporting more languages while preserving expressive speech controls.
User Experience & Interface Polish
- Latest Models First:Updated provider model lists so the newest Fish Audio, MiniMax, and Cartesia models are shown before previous generations.
- Smarter Expressive Controls:Emotion and pause controls now adapt more accurately to each model. Sonic 3.5 and Sonic 3.6 support the latest expressive controls, while incompatible formatting is automatically avoided for Fish Audio S2 models.
- Clearer Model Choices:Refreshed model names and descriptions to make it easier to compare high-quality, low-latency, multilingual, and expressive generation options.
Stability & Reliability
- Improved Provider Compatibility:Updated MiniMax Speech 2.8 routing and extended Cartesia generation settings—including speed, volume, and emotion—to Sonic 3.5 and Sonic 3.6.
- Safer Model-Specific Formatting:Added safeguards to prevent unsupported Smart Tags from being applied to newer Fish Audio models, reducing unexpected spoken instructions and formatting issues.
- Expanded Model Validation:Refreshed model-catalog, provider-routing, expressive-control, and generation coverage across the current Fish Audio, MiniMax, and Cartesia lineup.
new
Release
New Features & Major Enhancements
- Self-Serve Billing Portal:Added a dedicated Stripe Billing Portal flow for subscribed users. Users can now open the billing portal directly from the subscription page to view invoices, manage payment methods, and return back to VoiSpark after completing billing updates.
- Narration Model Upgrade to Sonic 3:Updated the narration backend to standardize Cartesia narration generation on Sonic 3. Older Sonic 2 and Sonic Turbo references have been removed from the active model catalog, giving users a cleaner and more expressive default narration experience with Sonic 3 support.
User Experience & Interface Polish
- Improved Subscription Management:Added a “Manage Payment” action to both desktop and mobile subscription panels. The button now includes clear loading states while the Stripe Billing Portal session is being created.
- Cleaner Model Experience:Simplified the Cartesia model surface by focusing on Sonic 3 as the supported narration model. This reduces outdated model choices and keeps narration templates aligned with the latest backend generation behavior.
- Better Project Creation Visibility:Narration creation flows now distinguish whether users start from onboarding, the home dashboard, or Voice Studio. This improves future product decisions around onboarding and project-start UX without changing the core creation flow.
Stability & Reliability
- Safer Billing Portal Access:Added backend safeguards so the billing portal is only available to users with an existing Stripe customer and subscription history. The system performs read-only customer lookup instead of creating new Stripe customers just to access billing management.
- Updated Sonic 3 Test Coverage:Refreshed narration preview, narration build task, narration history, and TTS API test references to use Sonic 3. This keeps automated and manual test scenarios aligned with the current narration model stack.
new
Release
New Features & Major Enhancements
- Comprehensive Onboarding System:We've built a complete onboarding experience that guides new users through their first narration project. The multi-stage flow includes voice source selection (clone, public voices, or celebrity voices), file upload with SRT support, and project creation. Users can now choose from existing cloned voices or record new ones directly in the onboarding flow, with smart voice recommendations based on their use case.
- Timeline Audio Editor:Introduced a powerful timeline editor for narration projects with visual waveform display, drag-and-drop paragraph reordering, and precise gap adjustments between segments. The editor features a transport bar with playback controls, zoom functionality (keyboard shortcuts supported), auto-scroll during playback, and a ruler for precise timing. Users can now see their entire narration project laid out visually and make fine-tuned adjustments.
- Smart Tags for Natural Speech:Added emotion tags to give users fine control over narration delivery. The system intelligently parses and serializes these tags, ensuring consistent formatting across different voice providers. Emotion tags let you add expressiveness to specific text segments.
- Export System with Configuration:Implemented a complete export pipeline with configurable options including subtitle generation (SRT format). Users can track export history, view latest export status, and download completed exports. The system supports unified concatenation and export paths with proper silence insertion between paragraphs.
- Voice Discovery by Category:Launched voice categories (formerly "occasions") that let users browse voices by context like Super Bowl, romantic, scary/creepy, and radio announcer styles. The system includes intelligent voice categorization and analysis, making it easier to find the perfect voice for any project.
User Experience & Interface Polish
- Refined Paragraph Toolbar:Redesigned the paragraph toolbar with smart tag insertion buttons, regeneration controls, and improved visual feedback. The toolbar now shows generation status with breathing animations and disables actions appropriately during processing.
- Enhanced Voice Selection:Improved voice selection across the platform with better filtering by gender, language, and use case. Added hover-to-play functionality on voice avatars, expanded language options, and integrated popular voice lists in the onboarding flow.
- Better Audio Playback:Enhanced the shared audio player with proper cleanup logic, paywall triggers, and 'ended' event handling. Implemented auto-pause during gap adjustments and improved seek functionality in the timeline editor.
- Zoom-to-Fit Functionality:Added zoom controls in the timeline editor with keyboard shortcuts, allowing users to quickly adjust their view to see the entire project or focus on specific segments.
Stability & Reliability
- Performance Optimizations:Implemented smart caching for narration previews and paragraph history, reducing generation times and costs. Improved voice loading performance with better state management using centralized stores. Enhanced scrolling behavior with custom scroll containers.
- Robust Error Handling:Added comprehensive error codes for voice limits, and public voice downloads. Improved error messages to be more actionable and user-friendly.
new
Release
New Features & Major Enhancements
- Narration Playground:We've introduced a powerful new narration feature that lets you create long-form audio content from documents. Upload your audiobooks, blog posts, podcasts, or conversation scripts, and our system will intelligently convert them into natural-sounding narrations. You can now manage multi-paragraph projects with ease, including paragraph-by-paragraph regeneration and history tracking.
- Voice Discovery & Collections:Finding the perfect voice is now easier than ever. We've added a new voice discovery page with occasion-based browsing (like Super Bowl voices, romantic voices, and more). You can now organize your favorite voices into custom collections, making it simple to manage and reuse voices across different projects.
- Celebrity Voice Expansion:Our celebrity voice library continues to grow! We've added NFL legends and Super Bowl stars including Tom Brady, Patrick Mahomes, Peyton Manning, Aaron Rodgers, and many more. We've also added Hollywood icons like Tommy Lee Jones and Willem Dafoe, plus new voice categories for scary/creepy and radio announcer styles.
- Enhanced Voice Library:The voice library has been completely redesigned with better filtering, search, and organization. You can now filter voices by gender, language, use case, and more. The new tab-based interface makes it easy to switch between public voices, your custom voices, and celebrity voices.
- Narration Preview for Free Users:Free users can now preview how their documents will sound before committing credits. The preview feature extracts a portion of your text and generates a sample, helping you choose the right voice and settings.
User Experience & Interface Polish
- Smart Voice Replacement:You can now bulk-replace voices across all paragraphs in your narration project with just a few clicks. This makes it easy to try different voices without manually updating each paragraph.
- Better Error Messages:Error messages are now more helpful and actionable, guiding you to solutions when something goes wrong.
Stability & Reliability
- Performance Optimizations:We've significantly improved the performance of voice loading and filtering. The voice library now loads faster, and searching through voices is more responsive.
- Caching Improvements:We've implemented smarter caching for narration previews and paragraph history, reducing generation times and costs when you regenerate similar content.
- Error Handling:We've strengthened error handling across the platform, especially for document processing, voice cloning, and narration generation, ensuring you get clear feedback when issues occur.
new
Releases
New Features & Major Enhancements
- Smart Tags for Expressive Speech:We’ve introduced "Smart Tags," an intelligent feature that analyzes your text and automatically suggests emotion and break tags. You can now easily insert, edit, and manage these tags—even on mobile—to make your narrations sound more human and dynamic.
- Expanded Voice Library:Our celebrity voice collection keeps growing! We’ve added a new "Stranger Things" category and introduced many new voices (including Adam West, Alan King, and others) to give you even more creative options.
- Preview Mode:Free users can now try out voice generation features more easily. The new Preview Mode allows you to generate speech with truncated text, so you can hear how a voice sounds with your content before upgrading.
- Enhanced Text Editor:We’ve upgraded our underlying text editor. Expect a smoother writing experience with better copy-paste handling, clearer character limit indicators, and improved support for long scripts.
User Experience & Interface Polish
- Mobile Experience:We’ve significantly improved the mobile interface. New dedicated drawers make it easy to edit emotion and break tags on smaller screens, and the overall navigation and sidebar have been refined for better touch usability.
- Streamlined Onboarding:Getting started is smoother than ever. We’ve refreshed the onboarding flow with clearer steps and smarter redirection, ensuring you land exactly where you intended after signing up.
- "Help You Pick" Improvements:Our voice recommendation tool is now faster and more intuitive, with better audio playback and filtering to help you find the perfect voice for your project.
- Visual Refresh:You’ll notice new icons, updated pricing tables with helpful tooltips, and a cleaner look across the landing pages and testimonials sections.
Stability & Reliability
- Performance Boosts:We’ve optimized how voice data loads and improved the connection stability for real-time features.
- Smarter Error Handling:We’ve improved error messages and validation, especially for downloads and text limits, so you’re never left guessing if something goes wrong.
- Backend Optimizations:Behind the scenes, we’ve strengthened our voice cloning and emotion tagging systems to be faster and more reliable, ensuring consistent performance even during high traffic.
new
Release
New Features & Major Enhancements
- Narration (block-level conversation): New paragraph‑based narration editor with per‑paragraph voice selection, word/character counts, paste cleanup, a global save shortcut, and automatic loading of your latest successful task. File upload supports progress, example projects help you get started, and downloads are more reliable. Access limits are clearly indicated with paywall prompts tied to your plan/credits.
User Experience & Interface Polish
- Narration flow: Clearer dialogs and toasts (deletion, project not found, ongoing tasks), improved errors, more accurate progress, and better defaults (prioritize your default voice, safer sorting). Voice avatar/readability tweaks, mobile empty states, and smoother loading.
- Voice discovery & filters: Cleaner advanced/provider tags, accent hierarchy fixed, consistent search/close behavior, and mobile fixes for the voice list.
- Navigation: Narration is now in the sidebar; “Coming soon” labels removed where live.
- Credit usage: Fixed chart X‑axis and tooltip times for accurate history.
Stability & Reliability
- Safer generation: Clear text‑length limits with helpful messages, support for empty paragraphs when needed, and more accurate progress so tasks don’t appear stuck. Access checks tie to your plan/credits with clearer feedback.
- Fewer failed runs: Longer processing timeouts reduce merge failures; interrupted jobs can resume on startup; temporary files are cleaned up automatically.
- Better downloads: Stable download links with format validation and clearer errors if something goes wrong.
new
Release
New Features & Major Enhancements
- Narration Service: Long-form narration is now available end-to-end. Create projects that respect your credit balance, support eligible voice cloning, and guard against empty inputs for smoother, reliable generation.
- Unified Product Pages: The Text-to-Speech, Voice Cloning, and Voice Changer pages have been rebuilt with clear hero sections, instructional visuals, and contact areas to tell a coherent story from first click to conversion.
- Guided Voice Discovery: “Help You Pick” and product tours now include richer questionnaires, locale-aware content, and saved progress so you can land on the right voice faster.
- Promotions & Subscriptions: New desktop/mobile paywall popups, a Free Gift campaign, refined benefit messaging, and updated credit displays make upgrade moments and plan value easier to understand.
- Instant Account Updates: Subscription changes and surveys now sync in real time via live updates, so what you see in the app always matches your account status.
User Experience & Interface Polish
- Smoother Onboarding: Tours remember your place, include multilingual examples, and guide you across TTS history, voice cloning, and Help You Pick without losing context.
- Clearer Voice Management: Visible clone limits, timely upgrade prompts, provider-specific notes, and refreshed celebrity galleries reduce guesswork during selection.
- Visual Refresh: New illustrations, avatars, and refined typography land alongside tighter spacing, layering, and dialog sizing for a cleaner, more consistent UI.
- Better on Mobile: Improved empty states, clearer filter recovery in voice lists, streamlined menus, and aligned subscription layouts make small-screen use effortless.
- Localized Messaging: More interface copy adapts to your locale with improved translation handling, so prompts and controls feel natural wherever you are.
Behind-the-Scenes Improvements
- Reliable Credits & Benefits: Strengthened payment and benefit logic, with extra logging and smart refreshes, keeps your balances accurate during plan changes.
- Stronger Audio Pipelines: Automatic audio metadata detection, clearer error codes for large uploads and edge cases, and sturdier exception handling improve success rates in cloning and voice changing.
- Stable Narration Runs: Input validation, paragraph synchronization, and safer handling of empty documents prevent wasted generation.
- Faster, Safer Releases: Updated infrastructure and deployment scripts, cleaner documentation, and ongoing code refactors reduce coupling and pave the way for quicker, more stable updates—without disrupting your workflow.
new
Release
New Features & Major Enhancements
-
Fresh Paywall Experience
: A new paywall on both desktop and mobile now clearly shows what each plan includes, helping you decide when an upgrade makes sense.-
Expanded Voice Library
: We've added a variety of new celebrity and character voice packs, so you have even more personalities to bring into your projects.-
Flexible Audio Downloads
: You can now export audio in additional MP3 and WAV quality levels, making it easier to share or edit your creations in the format you prefer.-
Smarter TTS Services
: Voice partners like ElevenLabs, Hume, and Minimax now support more languages and richer voice details, broadening the range of natural-sounding results you can generate.User Experience & Interface Polish
-
Richer Playback Controls
: New audio players, autoplay options, and download popovers give you quicker ways to preview, organize, and grab your audio.-
Easier Voice Discovery
: Filters and search tools now highlight accents, age ranges, and voice characteristics, helping you land on the perfect voice in seconds.-
Guided Text Creation
: Long-text warnings, improved cloning tips, and clearer onboarding messages keep you informed about limits and next steps as you work.-
Localized Messaging
: The interface now adapts to more locales with improved translation handling, so prompts and controls feel natural wherever you are.Behind-the-Scenes Improvements
-
Seamless Voice Sync
: Voice data, filters, and benefit rules now stay in lockstep between the app and backend services, cutting down on mismatched results.-
Optimized Integrations
: Updated dependencies and new server tooling keep voice synthesis, cloning, and analytics pipelines running smoothly even under heavy demand.-
Cleaner Codebase
: Documentation cleanups and internal refactors pave the way for faster future updates without disrupting your current workflow.new
improved
fixed
Releases
Major Features & Enhancements
Advanced Voice Filtering System
Find your perfect voice faster than ever with our new comprehensive filtering system. Filter voices by gender, use case, accent, and descriptive attributes across all voice providers (Minimax, OpenAI, Orpheus). Whether you need a professional narrator or a casual conversational tone, the right voice is just a few clicks away.
Seamless Voice Discovery
Experience lightning-fast voice browsing with our optimized voice list that smoothly handles thousands of voices. New intuitive filter components make discovering voices effortless, so you can focus on creating rather than searching.
Smart Subscription Integration
Stay informed about your voice access with clear VIP-only voice indicators and real-time subscription credit display on mobile. Know exactly what's available to you and track your usage at a glance.
Multilingual Text-to-Speech
Break language barriers with expanded multilingual greeting text support, making VoisPark more accessible for creators working across different languages and regions.
Mobile-First Experience
Enjoy a completely redesigned mobile interface with subscription management, streamlined navigation, and optimized layouts that work beautifully on any screen size.
User Experience Improvements
Instant Search Results
Search smarter with our new real-time search that delivers instant results as you type. No more waiting - find the voices you need immediately.
Rich Voice Information
Make informed decisions with detailed voice information including provider details, language specifications, and helpful usage tips displayed right where you need them.
Streamlined Voice Selection
Choose voices effortlessly with our redesigned selection interface. The process is now more intuitive and requires fewer steps to get you creating.
Clear Error Guidance
When issues arise, get actionable solutions instead of confusing error messages. Our improved error handling tells you exactly what happened and how to fix it.
Polished Interface Design
Enjoy a cleaner, more professional interface with updated typography, consistent spacing, and improved visual hierarchy that makes every interaction feel natural.
Performance & Reliability
Rock-Solid Stability
Experience fewer interruptions with our enhanced error recovery system that automatically handles issues behind the scenes, keeping your workflow smooth.
Superior Audio Processing
Upload audio files with confidence thanks to improved compatibility and processing that supports more formats while handling edge cases gracefully.
Faster Voice Loading
Get to work quicker with optimized voice libraries that load faster and present only the most relevant options for your needs.
Enhanced Account Management
Manage your subscription effortlessly with improved billing features, promotional code support, and streamlined account controls.
Strengthened Security
Create with peace of mind knowing your account and data are protected by enhanced security measures and improved access controls.
new
improved
fixed
Release
New Features & Major Enhancements
- Language-Based Voice Filtering: You can now filter our entire voice library by language, making it much easier to find voices that speak your preferred language.
- Enhanced Celebrity Voice Experience: We've added celebrity avatars and improved the voice cloning page layout to make working with celebrity voices more engaging and intuitive.
- Advanced Voice Organization: Voice models now feature descriptive tags, helping you understand their strengths at a glance and choose the best one for your needs.
- Enhanced Analytics Tracking: We've integrated PostHog for better error tracking and user experience insights, helping us continuously improve the platform.
User Experience & Interface Polish
- Improved Navigation: We've added breadcrumb navigation across model pages to help you understand where you are and navigate more efficiently.
- Better Theme Management: Your light/dark mode preference is now remembered more reliably, and theme switching is smoother across the entire app.
- Enhanced Voice Display: Voice items now show language information and helpful tooltips for at-a-glance understanding of each voice's capabilities.
- Cleaner Component Design: We've refined our header, footer, and navigation components for a more consistent and polished experience across desktop and mobile.
- Improved Error Handling: Error messages are now clearer and more helpful, with better recovery options when things go wrong.
- Enhanced Audio Controls: We've updated character limit handling and audio credit calculations to be more transparent and user-friendly.
Performance & Reliability
- Smarter Audio Processing: We've enhanced our audio processing with better WAV header validation and repair, reducing upload errors and improving compatibility with different file formats.
- Enhanced Memory Management: Our backend now features smart memory monitoring and alerting to ensure consistent performance even during peak usage.
- Improved Error Recovery: We've implemented retry logic and enhanced error handling throughout the platform, making the app more resilient and reliable.
- Better Voice Limit Management: Voice cloning limits are now enforced more accurately, with clearer error codes when limits are reached.
- Optimized Analytics: We've improved our analytics infrastructure with materialized views and enhanced data processing for faster insights and better user journey tracking.
- Enhanced Session Management: Session handling has been improved across both frontend and backend for more stable user experiences.
Load More
→