How to use Voice Factory TTS

Voice Factory TTS is a free tool that reads your text in a human voice and gives you an audio file (mp3, wav, m4a or ogg). No sign-up, and it can also produce a file where several people speak in turn, like a conversation.

At a glance

1. Add text — paste it in, or open a .txt file.
2. Pick a voice — choose one from the list and press ▶ to hear it.
3. Create — press the button and you get the audio plus a subtitle file.

The step indicator at the top shows how far along you are, so first-timers can simply follow it in order.

1. Add text

Paste into the Your text box, or use Open .txt. Korean files saved from Windows Notepad (ANSI / EUC-KR) are recognised automatically, so nothing comes out garbled. Up to 10,000 characters at a time. If you are not sure what to try, press Insert sample.

Leave the file-name box empty and the start of your text becomes the name. If you plan to make several files, naming them yourself — "lesson1_intro" and so on — makes them much easier to find later.

2. Pick a voice

By default the list shows only voices that can read Korean, with female voices grouped above male ones. Besides Korean-only voices, the ones marked "multilingual" also read Korean, with a slight foreign accent that gives a distinctive feel.

Press ▶ on any row to hear the beginning of your text in that voice. Compare a few, then settle on one. Pressing another row moves the selection there. On a first visit one Korean female voice is already selected.

3. Tone and effects

Tone sets speed, pitch and volume together. "News" is crisp and slightly quick, "Calm" is slow and low, "Soft" and "Cute" sit higher and brighter. After picking a tone you can fine-tune it with the sliders below.

Effects layer sound onto the finished voice: child and giant (pitch shifts), elder (a tremble), robot, phone (call-line sound), sharp (lifts the vocal range) and echo. Several can run at once — "Child + Echo" sounds like a fairy in a storybook, "Deep + Phone" suits a call scene. Child and Giant are opposites and cannot both be on.

4. Background music

"Calm music", "Rain" and "Waves" are generated on the server on the spot, so there is no copyright to worry about. Choose My music to upload your own mp3 instead (up to 15MB, kept 24 hours). Music volume defaults to about 40%; if it feels too quiet you can raise it without burying the voice. The music fades in and out at the start and end.

5. Choosing a file type

6. Getting your files

Finished files stack up in the Your files list. ▶ plays one, ↓ downloads it. The CC button is the subtitle file (SRT), timed to the sentences. Tick several and press "Download selected as zip" to get the audio and subtitles in a single zip.

Files are deleted automatically after 24 hours and are only visible in the browser that made them. Download anything you need to keep.

7. Fixing pronunciation

English abbreviations and symbols are sometimes read differently from what you expect. Register them under Pronunciation fixes — NH → 엔에이치, 3F → 3층 — and they are read that way. Subtitles keep the original spelling (NH, 3F), so they stay usable on video. The list is stored in this browser and is still there on your next visit.

Common questions

How much can I make per day?

50,000 characters per person (per browser) per day. Previews and finished files both count against it, and it refills at midnight (KST). What is left is shown in the Create panel.

Yesterday's file has disappeared.

To save server space, anything older than 24 hours is deleted automatically. Download what you need right after making it.

I cannot see my files on another device.

In exchange for not requiring an account, files are tied to the browser. Download them on the device that made them and move them across.

Can several people speak in one file?

Yes. Switch to Script (several) at the top right of the text box. Dialogue audio explains it in full.

Use cases