HeyGen Voice
HeyGen Voice is the first in-house voice model from HeyGen, the AI video company, launched on 9 October 2026. HeyGen says it debuted at the top of Artificial Analysis' voice leaderboard and includes it at no extra cost in every HeyGen video. On fal it reads a script in one of 24 library voices, or clones a voice from a reference recording, using up to its first three minutes, across 39 languages, and returns 44.1 kHz WAV. It costs $0.03 per thousand characters on fal, the same $30 per million characters HeyGen lists for its own API. A cheaper Turbo tier sits beside it.
Anvisha Pai, Co-founder & CEO, Voyager
HeyGen · released
›What we asked for
Hi, I'm HeyGen Voice, the voice model from HeyGen, the AI video company. I can read in one of HeyGen's library voices or clone a voice from a short recording. Here is the same script every voice on this site reads, so you can compare us.
heygen/heygen-voice
voice: Cecily
returned 15.2 seconds of wav
At a glance
- Price
- $0.03 per thousand characters on fal, the rate HeyGen lists for its API; our four scripts cost about three cents.
- Voices
- 24 named library voices on fal, Jonah by default; we used Cecily, fal's playground pick.
- Cloning
- Pass a reference_audio_url and it clones that speaker, using the first three minutes; with Turbo, the only voice models on this site that clone through fal.
- Languages
- 39, detected from the text unless set.
- Turbo tier
- HeyGen Voice Turbo takes the same request at $0.004 per thousand characters.
Other HeyGen Voice models: HeyGen Voice Turbo. Compare all 2.
Anvisha's take
We have not scored the four readings by ear; press play and judge the voice yourself. What we measured: HeyGen Voice read the four scripts in 12 to 23 seconds of audio each for three cents across all four. Its names-and-numbers reading, 20.6 seconds, and Turbo's, 20.5, are the shortest on the site; a short read there can mean quick speech or a figure read more compactly, and we have not checked it word by word. Its first request took 23 seconds to return, longer than any reading from the four September models, and the next three took five to seven. HeyGen has none of the voices the other pages use, so it read with Cecily, the voice fal's playground preselects, at default expressiveness.
HeyGen Voice real output: the same 4 scripts every model gets
Every voice model on this site gets the same 4 scripts, so you can compare like with like. These are the readings HeyGen Voice returned, one try each, nothing re-rolled or retouched. Generated on . The 4 readings took 5 to 23 seconds each and cost $0.0292 in total.
Pacing, warmth and whether the pauses land where a voice actor would put them.
›What we asked for
Meet the Ember mug. It keeps your coffee at exactly the temperature you choose, for up to ninety minutes, and it tells your phone when it's ready. Warm from the first sip to the last. Ember. Coffee, on your terms.
heygen/heygen-voice
voice: Cecily
returned 13.7 seconds of wav
Steady narration over a long paragraph, clean list rhythm and a natural close.
›What we asked for
A pour-over works in four steps. First, rinse the paper filter with hot water so it doesn't taste of paper. Second, add the grounds and pour just enough water to wet them, then wait thirty seconds while they bloom. Third, pour the rest of the water in slow circles. Fourth, wait about three minutes for it to drip through. That's it: a cup that tastes like the coffee, not the machine.
heygen/heygen-voice
voice: Cecily
returned 23.0 seconds of wav
Range: excitement, then a hushed aside, from the words alone with no tags.
›What we asked for
We did it! We actually did it! I can't believe it's finally over. Okay. Okay, deep breath. Let's not tell anyone until Monday, alright? Just... let me have this one quiet night.
heygen/heygen-voice
voice: Cecily
returned 11.6 seconds of wav
Reading numbers, times, currency, an email address and two hard names correctly.
›What we asked for
Your order ships on the 24th of September 2026 and arrives by 9:15 a.m. The total is $1,249.99, including VAT. Questions? Ask for Siobhan Nguyen or Ravi Parikh on 0800 555 0199, or email help@moda.app.
heygen/heygen-voice
voice: Cecily
returned 20.6 seconds of wav
How the cost under each picture was worked out: fal's HeyGen Voice rate of $0.03 per 1,000 characters (fal model page and pricing API, read 2026-10-22) for the 213 characters sent; fal does not return the charge.
Compare HeyGen Voice
HeyGen Voice pricing
| Where | Unit | Price | Source |
|---|---|---|---|
| fal | 1,000 characters | $0.03 | fal model page and pricing API, read |
| HeyGen API | 1,000 characters Listed as $30 per million characters with no tiers; HeyGen's launch release called the model free within its platform and API. | $0.03 | HeyGen Voice page, read |
Is HeyGen Voice free?
HeyGen says HeyGen Voice is free in the HeyGen app, included in every video at no extra cost; its launch release also called it free in the API, while its voice page lists API use at $30 per million characters. fal charges per character.
- •HeyGen app: included in every HeyGen video.
- •HeyGen API: $30 per million characters on the voice page.
- •Professional voice clone: a paid add-on in HeyGen.
- •fal: $0.03 per thousand characters.
HeyGen Voice page, read
Where to use HeyGen Voice
- HeyGen · In the HeyGen video app, with instant and professional voice clones.
- HeyGen API · REST and streaming endpoints.
- fal · $0.03 per thousand characters, library voices or a clone from a recording.
- Voyager · Runs inside Voyager, the agent for creative work, with your own fal key.
Voyager
Run HeyGen Voice inside Voyager
Voyager is an agent for creative work that runs every model in this reference, HeyGen Voice included. Brief it, and it picks the model, makes the thing and hands back something you can edit. Create an account to get started.
What we noticed running HeyGen Voice
- •The first request of the run took 22.9 seconds to return; the other three took five to seven. See the ad read reading.
- •20.6 seconds; with Turbo's 20.5, the shortest readings of this script on the site. We have not checked whether every figure was read in full. See the names, dates and prices reading.
- •11.6 seconds, the same length to the hundredth as HeyGen Voice Turbo's; only Kokoro, at 11.5, read it shorter. See the emotional line reading.
HeyGen Voice limits and API parameters
| Limit | Value |
|---|---|
| Script length | 5,000 characters a request. fal API reference, read |
| Clone reference | Only the first three minutes of the recording are used. fal API reference, read |
| Output | 44.1 kHz WAV, with the duration in the response. fal API reference, read |
›Every setting, for developers
fal (heygen/heygen-voice)
fal API reference, read
| Parameter | Type | Values | Default |
|---|---|---|---|
| text | string | The script, up to 5,000 characters Required. | |
| voice | enum | Jonah, Nadine, Tabitha, Soren, Elowen, Vincent, Jocelyn, Marielle, Corinne, Lucian, Kieran, Hugh, Noelle, Bram, Ewan, Iona, Cecily, Simon, Weston, Felicity, Gareth, Rory, Meredith, Sabrina Ignored when reference_audio_url is set. | Jonah |
| reference_audio_url | string | A recording of one speaker to clone Only the first three minutes are used. | |
| expressiveness_boost | number | 0 to 1 | 1 |
| language | enum | 39 language codes, from ar to zh Detected from the text when omitted. | |
| enable_safety_checker | boolean | true, false Only authorised callers may turn it off. | true |
What HeyGen Voice will not say
HeyGen's content moderation policy bars fraud, harassment, sexual content, child-safety violations, hate and misinformation, and requires explicit consent before making an avatar of a real person. It does not mention voices or voice cloning, so the consent rule for a cloned voice through fal's reference_audio_url is not spelled out there.
Alternatives to HeyGen Voice
Frequently asked questions
What is HeyGen Voice?
HeyGen's first in-house text-to-speech model, launched on 9 October 2026. It reads in library voices or clones a voice from a recording, in 39 languages, and is built into HeyGen's video app.
How much does HeyGen Voice cost?
$0.03 per thousand characters on fal, and $30 per million characters on HeyGen's API, read on 22 October 2026. It is included in HeyGen videos at no extra cost.
Can HeyGen Voice clone a voice?
Yes. On fal you pass a recording of one speaker and it uses the first three minutes. Only clone a voice you have the owner's permission to use.
How this page is made
Prices, limits and settings are read from the linked vendor and host pages on the dates shown. Every picture was generated by us through fal with the request shown under it, and the original files are kept unedited. A generation that failed is shown as a failure. Model and vendor names are used to identify the products; no vendor imagery appears on this page.
