What it does
TTS Lines is a free text-to-speech editor that runs in your web browser. You paste a script, it splits the text into sentence-level lines, and you generate audio for each line. If one word or name sounds wrong, you edit and regenerate only that line instead of producing the whole recording again.
In the free editor, speech is generated on your device. The first time you use it, the site downloads a neural voice model from Hugging Face into your browser, and the model then runs locally through WebAssembly. Your script and the generated audio are not sent to a speech service in this free workflow, according to its privacy page. It does not use your browser's built-in system voices.
When the lines sound right, you can export one joined recording, separate clips for each line, and a caption file. TTS Lines is made by AudioMultiCut under the Endless Apps brand. A paid Pro plan with cloud voices and saved projects has been announced, but it could not be bought when we checked.
Who it's for
YouTubers, course creators and writers who need short voiceovers they can correct one sentence at a time.
Key features
- Sentence-level editing: each sentence becomes a line you can edit, move, regenerate or add to
- A main voice for the script, with a different speaker on any line for dialogue
- Per-line speed (0.8× to 1.2×) and the pause after each line (none, 0.3, 0.5, 1 or 3 seconds, or a custom value)
- Local preset voices: the editor lists ten for English (labelled F1 to F5 and M1 to M5), and Hebrew has four
- 32 languages in the language menu, with auto-detect. Hebrew uses a separate model that adds pronunciation to unpointed text
- Exports: one joined MP3 (128 or 192 kbps) or WAV, a ZIP with one MP3 per line, a single-line MP3, and an SRT caption file
- SRT cues timed from each generated line and the pauses you set
- No account needed for the free editor. Your draft is saved in the browser
Based on TTS Lines' website, pricing page, terms, privacy page and guides, checked October 9, 2026. Features change, so check the official site.
Pros and cons
- Free to use, with no account, according to its pricing page
- The free local voices generate speech on your device, so your script is not sent to a speech API
- Line-level regeneration makes small corrections quick
- Separate clips per line suit video editing, where timing often changes after narration is recorded
- Exports MP3, WAV, ZIP and SRT on the free plan
- Clear, published notes on its own limits and on the open-source models and licences it uses
- The free voices are a small set of presets. There is no voice cloning, and no detailed control over emotion or emphasis
- There is no custom pronunciation dictionary in the free editor, so difficult words may need a phonetic spelling
- The first use downloads a large model, and speed depends on your device and memory. The site says low-memory phones may struggle and suggests a desktop browser
- Local use is not fully offline: the site and model need to load, and clearing browser data can mean downloading the model again
- Lines have a length limit, and Hebrew digits must be written as words
- The free editor keeps one draft in the browser. Clearing browser data deletes it, so keep your own copy of the script
- SRT captions are timed per line, not per word, and need retiming if you edit the audio elsewhere
- The terms don't say clearly whether you can use the audio commercially (see the pricing notes)
- Pro features such as cloud voices and saved projects can't be used yet
Practical use cases
- Narrating a short YouTube explainer or tutorial, then exporting one clip per line to place on a video timeline
- Fixing a mispronounced name, acronym or number in one sentence without re-recording the script
- Making a simple two-voice dialogue for a lesson or social video by setting the speaker line by line
- Producing a draft voiceover for a course module or a client review before deciding on a paid voice
- Generating starter SRT captions that match the exported recording
- Voicing scripts you would rather not upload to a cloud service, as long as you use the free local voices
Pricing
Pricing checked October 9, 2026. Prices and plan details can change.
- Free: $0, no account. Local voices, line-by-line editing, separate speakers, speed and pauses, MP3, WAV, ZIP and SRT exports, and one browser draft
- Pro (announced, not yet available): $15/month plus tax, with 100,000 cloud speech characters per paid billing cycle, cloud voices, projects synced across devices, folders, templates and 10 saved versions per project
Pro can't be bought yet. The pricing page says "Pro checkout is being set up", and the Projects page says saved projects will open when Pro launches. According to the terms, every cloud generation and regeneration will count towards the character allowance, and unused characters won't carry over. Payments are planned through Paddle, and the seller may show as Melioris AI.
Privacy and rights: the privacy page says Pro cloud voices will send the selected line's text to Google, through Vercel AI Gateway, for generation. Local voices don't send your script anywhere. The free voices use Supertone's Supertonic 3 model under the OpenRAIL-M licence, which the terms incorporate. That licence says the licensor claims no rights in the audio you generate, subject to its use restrictions. The TTS Lines terms don't explicitly address commercial use, so check the terms or contact the developer before using the audio in paid work. Check TTS Lines' official pricing page for current details.
How it compares to alternatives
Most cloud text-to-speech tools send your whole script to a server and return one finished file, usually with a large voice library and a monthly character allowance. TTS Lines works differently. The free voices run in your browser, cost nothing per line, and are built around fixing and exporting individual sentences. The trade-off is a smaller set of preset voices, fewer controls and a reliance on your device's performance.
If you need a wide voice library, voice cloning or long-form production tools, ElevenLabs and Murf AI are fuller options, and our Murf AI vs ElevenLabs comparison covers how those two differ. Whichever tool you use, our guide to writing a script for AI voiceover will help the result sound clearer.
Editorial view: TTS Lines is a sensible free option for short voiceovers where you want to fix individual sentences quickly and keep your script on your own device. Its per-line clips and SRT export fit a video-editing workflow well. It is a narrow tool rather than a full voice studio: the voices are presets, the first load can be slow, and the paid Pro plan is announced but not yet available. Treat Pro features as unavailable until checkout opens, and check the licence terms before using the audio commercially.
This overview is based on TTS Lines' published product information, pricing page, terms, privacy page and guides, checked October 9, 2026. It does not describe hands-on testing.
Want to try it yourself?
The best way to judge fit is to test it on a real piece of your own work.