1.2.1

1.2.1

• Added Previous Page alongside Next Page in Generation History, Voice Library and Pronunciation Dictionaries on Windows and Mac.
• Page navigation is available only when there is a page to visit. Refresh returns to the first page.
• Failed page requests retain the current results and navigation so you can retry.
• Changing a search or switching between account and shared voices starts a new set of results after Refresh.
• Updated the manuals and bundled Version History, including the History Read API-key permission needed for Generation History.

Windows, Apple Silicon Mac and Intel Mac downloads are available below. Existing preferences, API keys, projects and recordings are preserved by the signed updater.

1.2.0

Voice Library and Keyboard Improvements

• Optional automatic previews when selecting voices. The saved checkbox is beside Preview / Stop.
• Preview / Stop toggles playback; Enter on Windows or Return on Mac does the same from the voice list.
• Rename voices owned by your account without changing their voice IDs or recordings. Voices Write permission is required.
• More official speech-tag examples, with multi-letter navigation in Windows tags and voice dropdowns.
• Fixed Tab and Shift-Tab navigation through the Mac speech-tag picker.

Windows and both signed, notarized Mac downloads are available. Existing preferences, keys and recordings are preserved during updates.

1.1.1

ElevenLabs Speech Generator 1.1.1

• Added Version History to the Help menu, available offline.
• The Mac app opens with keyboard focus in the main text editor when the selected mode has one.

Previous Release: 1.1.0

• Optional long-text speech generation uses sentence-aware splitting, assembles one WAV per variation and can recover completed parts.
• Voice cloning shows readable upload progress and the final result.
• Windows context help provides specific explanations for cloning controls, Preferences and other dialogs.
• The Mac variations field accepts a typed number.

1.1.0

What's New

• Optional long text-to-speech supports up to 500,000 characters. Enable it in General preferences. The app splits at sensible boundaries within the selected model's limit and assembles one WAV per variation.
• Identical unfinished requests can reuse verified completed parts. Each new part is a paid request; a request accepted by ElevenLabs can still incur a charge if interrupted. Review voice continuity at joins before publishing long recordings.
• Voice cloning shows readable upload progress, server processing and the final result. Close and Cancel Upload have separate Windows mnemonics.
• Windows F1 help gives specific explanations for cloning controls, Preferences and other dialogs.
• Mac variations can be typed directly, and window titles show the current character count.

Downloads

Choose the Windows ZIP, Apple Silicon Mac ZIP or Intel Mac ZIP. The Windows .sig is used by the automatic updater; it is not needed to unpack and run the app manually. Mac downloads are Developer ID signed and notarized.

Existing preferences, API keys, projects and recordings are preserved by the updater.

ElevenLabs Speech Generator 1.0.0

ElevenLabs Speech Generator 1.0.0

Initial release of an accessible native speech and audio application for Windows and Mac, by Andre Louis.

Included workflows

• Text to speech and multi-voice dialogue with voice settings and a speech-tag picker.
• Transcription to text and subtitles, plus alignment of an existing transcript to audio.
• Voice conversion and background-noise isolation.
• Voice design and remix previews, a searchable voice library and cloning tools.
• Automatic dubbing of audio or video, with saved jobs and resume without creating another paid project.
• Pronunciation dictionaries, generation history and reusable projects.
• Output-device selection, optional sequential playback and a readable credit balance.
• Native keyboard controls, screen-reader-friendly dialogs and F1 help for the focused control.
• Signed updates, startup update checks and a bundled HTML manual.

Downloads and setup

Windows: download ElevenLabsSpeechGenerator.zip, extract the whole archive into a writable folder and run the executable. Windows 10 or later and .NET Framework 4.8 are required. Keep the accompanying files beside the executable. The updater retrieves the separate .zip.sig automatically.

Mac: choose the Apple Silicon ZIP for an M-series Mac or the Intel ZIP for an Intel Mac, extract the app and move it to Applications. Both require macOS 14 or later and are signed and notarized.

An ElevenLabs account, sufficient credits and your own API key are required. Use Preferences or Settings to save the key, set your output folder and test access. Enable only the permissions needed for your workflows; User read access enables the balance display. Service charges and available models depend on your subscription.

Automatic dubbing spends credits when a project is accepted. Stopping local monitoring does not cancel the remote job or its charge; resume the saved project to retrieve its result.

Only clone or process voices and recordings you have the necessary rights and consent to use. Review generated transcripts, translations and subtitle timing before publishing them.

For music and sound effects, see the companion Music and Sound FX Generator.

Software catalogue | Donate