ElevenLabs v4 Text to Speech

ElevenLabs v4 text to speech, no ElevenLabs account: paste a script, pick one of 20 voices on Eleven v4 or v4 Turbo, get an MP3 in seconds. From 1 credit.

  • Eleven v4
  • v4 Turbo
  • from 1 credit
Voice model
Voice
Text to read0/3000
20 credits per 1,000 characters

Eleven v4: 20 credits per 1,000 characters, rounded up to a whole credit. Charged once when you click; a failed voice-over returns its credits.

Your voice-overs

Sign in to keep your voice-overs. Your latest 20 stay here to replay and download.

Real voice-overs from this tool

Recorded on 2026-10-02: each is one real request sent by this tool's server code, uncut, with the text that was read, the voice, the length and the credits it costs.

  • “[excited] Meet the mug that keeps your coffee hot for six hours. [whispering] Double-walled steel, and a lid that never leaks.”

    Eleven v4 · Aria · 7s · 126 characters · 3 credits · 01a0fa59-d07b-7362-9643-5eac459841d3

  • “The lighthouse had been dark for forty years. Tonight, for the first time, someone climbed the stairs with a lantern.”

    Eleven v4 · George · 8s · 117 characters · 3 credits · 01a0fa59-df91-7542-8a17-b9ffcb018bb7

  • “Step one: open the settings menu. Step two: choose Export, then pick MP4. Your video is ready in under a minute.”

    Eleven v4 Turbo · Jessica · 9.4s · 112 characters · 2 credits · 01a0fa59-ecfe-7e30-a603-7cf6a23976b0

  • “欢迎来到 Cicadas。输入一段文字,选择音色,几秒钟后就能得到配音。”

    Eleven v4 · Aria · 6.4s · 36 characters · 1 credit · 01a0fa59-fb7f-7df1-b6fd-a7e4e02a207d

About ElevenLabs v4 on Cicadas

Eleven v4 is the newest ElevenLabs speech model; fal has offered it since 2026-09-28, with a faster v4 Turbo beside it. Paste up to 3,000 characters (1,000 in Chinese, Japanese or Korean), pick a voice, and add tags such as [excited] or [whispering] where the delivery should change. Cicadas returns an MP3 you can replay and download. The price shows before anything is charged (20 credits per 1,000 characters, 10 on v4 Turbo), and a failed voice-over returns its credits. In our test on 2026-10-02, scripts of 112 to 126 characters came back as 7 to 9 seconds of speech, and a listener that was not shown the scripts wrote down every word correctly; in eight runs on this page a voice-over was ready to play 8 to 12 seconds after the click.

The other voice here is Gemini TTS. A finished MP3 can narrate a faceless video, be fitted to a video on lip sync, or play over your photo on the subway singing page. How to write [excited], [whispering] and other tags, with our four recorded voice-overs: Eleven v4 audio tags guide.

Why use Cicadas for ElevenLabs v4

Eleven v4 and v4 Turbo on one page

Switch between the full model and Turbo, which fal lists at half the price, and hear the same script on both. Each reads the same 20 preset voices.

Direct the delivery with tags

Write [excited] or [whispering] inside the text and v4 changes its delivery from that point. In our test the listener described the tagged line as enthusiastic, then whispered, without being told the tags.

Priced per character, kept in your account

20 credits per 1,000 characters on v4 and 10 on Turbo, rounded up to a whole credit: a 120-character line costs 3 credits on v4 and 2 on Turbo. Your latest 20 voice-overs stay on this page and in your Studio library.

How to use ElevenLabs v4

  1. 01

    Paste your script (up to 3,000 characters, or 1,000 in Chinese, Japanese or Korean) and pick Eleven v4 or v4 Turbo.

  2. 02

    Choose one of the 20 voices, and add tags such as [whispering] where the delivery should change.

  3. 03

    Check the price, generate, listen, then download the MP3.

How do audio tags work in ElevenLabs v4?

You type a tag in square brackets inside the script, and the delivery changes from that point. Our test line began with [excited] and switched to [whispering] halfway: 126 characters read by the Aria voice came back as 7 seconds of speech. A listener that was not told about the tags described it as enthusiastic, then whispered.

What file do I get, and where are my voice-overs kept?

You get an MP3 that you can replay and download on this page. Your latest 20 voice-overs stay on the ElevenLabs v4 page and in your Studio library. In our test on 2026-10-02 the four recorded files were 44.1 kHz MP3s, a voice-over made through the page downloaded for its owner, and a signed-out request for it was refused.

ElevenLabs v4 or Gemini TTS: which voice page should I use?

Use this page when you want to place delivery tags inside the script, and Gemini TTS when you would rather describe the delivery in plain words. ElevenLabs v4 here has 20 preset voices and returns an MP3; Gemini 3.8 Flash TTS has 30 voices and returns a WAV. Both accept scripts of up to 3,000 characters, with lower limits for Chinese, Japanese or Korean.

ElevenLabs v4 prompt ideas

Copy one into the prompt box above and adapt it to your shot.

  • “[excited] Meet the mug that keeps your coffee hot for six hours. [whispering] Double-walled steel, and a lid that never leaks.”

  • “The lighthouse had been dark for forty years. Tonight, for the first time, someone climbed the stairs with a lantern.”

  • “Step one: open the settings menu. Step two: choose Export, then pick MP4. Your video is ready in under a minute.”

ElevenLabs v4 FAQ

Tested on Cicadas · reviewed 2026-10-02

Recorded on 2026-10-02. Four direct requests in this tool's request format (Eleven v4 with Aria and George, v4 Turbo with Jessica, one in Chinese): scripts of 36 to 126 characters returned 6.4 to 9.4 seconds of 44.1 kHz MP3, 3 to 5 seconds after each request. A listener model that was not shown the scripts transcribed all four word for word and described the tagged line as enthusiastic, then whispered. Eight voice-overs made by clicking through this page on a local build were each charged once (2 credits), could be played 8 to 12 seconds after the click, downloaded as MP3 for their owner and returned 401 to a signed-out request. Each request goes to the provider's queue once and is checked on every second, so a slow one does not hold up other jobs. The same request sent twice was charged once, a mismatched price was refused, and a refused, failed, unusable or overdue provider result returned the credits. In a separate check all 20 voices returned audio, and a made-up voice name was refused. Prices follow fal's list price (USD 0.08 per 1,000 characters, 0.04 on Turbo) at 250 credits per USD. Long scripts (direct requests, same day): 3,000 English characters returned about 3 minutes 10 seconds of speech after 47 to 48 seconds on both models; a numbered list of 2,876 Chinese characters returned about 8 minutes 13 seconds after 169 seconds (Turbo) and 219 seconds (v4), and the last 25 seconds matched the end of the script. Through this page, an original Chinese story of 1,018 characters on v4 Turbo cost 11 credits and came back after 40 seconds as 3 minutes 32 seconds of speech whose start, middle and end matched the script. One request is limited to 3,000 characters, with Chinese, Japanese and Korean characters counted as three (1,000 per request), so that a script is three to four minutes of speech and well inside the provider's five-minute limit. Voice cloning, multi-speaker dialogue and longer scripts are not offered here.

Updated Nov 14, 2026 · How we test

Related guides