- NewVoice Clone Pro is now multilingual: clone voices in English, French, German, Italian, Portuguese and Spanish.
- NewChoose from three English models (English, English 1, English 2) for different voice styles.
- NewNew high-quality French model for richer, more natural French speech.
- NewDownload only the languages you need from the new "Manage models" window.
- NewEach language comes with ready-made preset voices you can download with one click or all at once.
- NewSave your cloned voices and reuse them anytime without uploading the audio again.
- NewLong text is now split into smart chunks at natural sentence breaks, so long scripts sound smooth and never cut off.
- NewBrand new UI/UX: a cleaner, more modern look that is easier to use.
- ImprovedClearer progress bars and status messages while models load and audio is generated.
- ImprovedAPI upgrade: new Voice Clone Pro endpoints to list, download and load language models and voices from your own apps.
- ImprovedAPI upgrade: save and use cloned voices through the API.
- ImprovedAPI upgrade: OpenAI-compatible speech endpoint, so apps built for OpenAI text-to-speech can work with MusaTTS.
- ImprovedAPI upgrade: built-in API documentation (OpenAPI) with clearer error messages that tell you exactly what to do next.
- ImprovedImproved speed and stability across the app.
- FixedMany small bug fixes and polish.
Changelog
What’s new in Musa TTS
Every release, from new voices to API upgrades. Lifetime license holders get all of them free.
- FixedBugs Fxed
- NewVoice Store Added
- ImprovedEnhanced Voice Clone Pro
- ImprovedSpeed-optimized
- Fixedfixed The Issues of hanging on to low-end cpus
- ImprovedVoice Clone Pro Enhanced
- NewVoice Clone Pro Added
- NewEnterprise-grade voice cloning with just 5–10 seconds of reference audio
- NewNo base voice selection needed — upload reference audio, type your text, and generate
- NewTemperature control slider to fine-tune voice variation and naturalness
- NewPause markers — insert natural pauses directly in your text:
- NewWeak (0.15s), Clause (0.3s), Sentence (0.6s), Paragraph (1.0s), or custom durations
- NewSay-As markup — control how text is pronounced:
- NewCardinal, ordinal, digits, characters, dates, times, phone numbers, units, and fractions
- NewPre-load engine button so your first generation is instant
- NewReal-time progress bar during model loading and generation
- NewFree users get 2,000 characters/day to try it out
- NewBuilt-in REST API to control MusaTTS from any app, script, or workflow
- NewEndpoints: health check, list voices, TTS synthesis, voice blending, and Voice Clone Pro
- NewAPI key authentication with one-click generation and regeneration
- NewConfigurable port (1024–65535)
- NewInteractive documentation with ready-to-copy examples for Python, cURL, and JavaScript
- NewStart/stop the server anytime from the new API Settings page
- NewNew Voice Clone Pro page with drag-and-drop reference audio upload and audio preview
- NewNew API Settings page with enable/disable toggle, key management, and live server status
- NewEngine status indicator shows whether the Voice Clone Pro model is loaded and ready
- NewClone any voice by uploading a short audio sample of the speaker
- NewType your text, pick a base voice, and generate speech that sounds like the uploaded voice
- NewFilter voices by language for quick selection
- NewPreview your reference audio before generating
- ImprovedAdjust speech speed with a simple slider
- NewAudio plays automatically after generation
- NewSave your cloned audio as a WAV file
- NewAudio-to-Audio Clone
- NewTransform any audio file to sound like a different speaker
- NewUpload your source audio and a reference voice — MusaTTS handles the rest
- NewSupports WAV, MP3, FLAC, OGG, and M4A files
- NewPreview both files before cloning
- NewSee real-time progress while your audio is being converted
- NewFree & Pro Plans
- NewFree users get 1,000 characters per voice clone, up to 3,000 characters per day
- NewFree users get 1 audio-to-audio clone per day
- NewPro users get 5,000 characters per voice clone (adjustable up to 20,000)
- NewPro users get 10,000 characters per TTS generation (adjustable up to 50,000)
- NewPro users get unlimited daily voice clones and audio clones
- NewChange your per-generation limit anytime from the settings on each page
- ImprovedImproved Upgrade Prompts
- NewWhen you hit a limit, a clear dialog shows exactly what you've used across TTS, Voice Clone, and Audio Clone
- ImprovedOne-click upgrade button takes you straight to the license page
- FixedFixed an issue where deactivating a license could leave the app stuck in Pro mode
- FixedFix Issue Of Index Out of bound