A clear guide to the studio: voiceover, voices and creating your own voice, voice conversion, transcription, export, plans and API. Pick a topic on the left. For integrations there is a separate API documentation.
HancVoice turns text into natural speech with a neural network. You can voice scripts, articles, books and ads with system or your own voices, and transcribe audio and video into text.
The main studio screen. Here you enter text, choose a voice and get audio.
.txt or .md.Choose a voice and language at the bottom. “Smart intonation” assigns emotions per sentence automatically; you can set an overall mood and speech speed.
The “Voice” button plays the result in the browser (free trial). The “Export” button creates a finished file and charges characters for the whole text once — more in the “How characters are counted” section.
You don’t have to type or paste text: the “File” button loads .txt and .md (encoding is detected automatically), plus .fb2 and .epub books — chapter text appears in the editor. For voicing whole books, the dedicated “Audiobook” section is more convenient.
Each generation differs a little — that's normal for a neural network. You can pin a version you like, and then the sound stops changing.Fine-tune pronunciation over pieces of text — the “How to read” panel in the editor.
The AI brings “raw” text to a pronounceable form: it expands numbers, abbreviations, law references and acronyms.
“Auto” determines the type itself. Preparation runs on a button, not by itself.
Smart preparation is available on a paid plan. Without it, the voiceover still normalizes the text at a basic level.More than 326 ready voices in 24 languages — male and female, with different characters. 10 languages come with a quality guarantee, plus 14 more in beta via auto mode (all languages).
Voices you create appear in the same list and in the editor's dropdown. Manage them via the “⋯” menu on a voice.
Create a private voice model from a sample recording.
Transfers the delivery — an actor’s performance and intonation — onto the voice you choose. A paid-plan feature (Starter and up); the section is marked “beta”.
The finished result can be refined in the editor — equalizer, “Make cleaner”, backgrounds (see Sound and effects).
A marketplace of authored voices: an author publishes their voice, a buyer gets access to published voices. The section is getting ready to launch — we're opening early access to authors.
It's the next layer after system and private voices: the author fills out a voice card (description, tags, demo), confirms rights, passes moderation — and the voice becomes available to other users.
Sales and the author's share are tracked in the account. At launch, payouts are by a manual request after checking your details and, depending on your tax status, the required documents. Exact terms and thresholds are in the “Payout rules” document. Automated payouts are being worked on as a separate stage.
A voice is personal data. You may publish only your own voice or a voice with documented consent and rights — see “Voice rights and privacy”.Voice an entire book chapter by chapter: upload a file and get an audiobook in the voice you choose. Voicing runs in the background — you can close the page.
Characters are charged only for what is actually voiced: the whole book on the first run, and only unvoiced or edited chapters afterwards. If voicing is interrupted, characters for unfinished chapters are refunded automatically. Audiobooks are available on paid plans.
Every chapter has a pencil button: open the text, fix typos, insert pauses (‖) or run “Smart prep”; mispronounced words are fixed via the pronunciation dictionary (maintained by support). After saving, the chapter is marked “edited” and can be re-voiced on its own — the cost is shown right in the row.
While a chapter plays, the passage being spoken is highlighted in the text — you read along and immediately see where something sounds off. Click any place in the text to jump the player straight there.
If only one sentence bothers you, there is no need to re-voice the whole chapter: each passage has a “Re-voice passage” button. The new audio is spliced precisely into the existing file, the rest of the chapter stays untouched — and only the characters of that passage are charged.
Besides individual chapters you can download the complete book as one file: MP3 (a single continuous file) or M4B — the audiobook format with chapter markers, which players show as a table of contents and use to remember your position. The file is assembled on first download; chapters added later require assembling it again.
After voicing, a robot listens to each chapter and compares it to the text. If something sounds off, the chapter row shows “⚠ suspicious spots” — clicking a spot jumps the player right there.
Chapters can be renamed, deleted, reordered and added manually — the buttons live in the chapter edit panel.Audio and video into text — with timecodes. Transcription is available from the Pro plan and up (it is not included on Free or Starter); speaker labels and summaries come with the Business plan.
The “Sound” block in the editor — final processing before export.
How to get the finished audio file.
A very long text (a book, a large script) is exported in the background: you can close the tab, and an email arrives when it's ready. Progress is visible in the editor.
From 5,000 characters: the text is split into chunks, the timbre stays stable across the whole document and loudness is levelled at the end — the same way our blog articles are voiced. Markup, emotions and intonation are not applied in this mode; the export runs in the background and is available on paid plans.
Re-downloading the same finished file is free. Characters are charged once when the file is created.Details on finished exports: where to find them, how long they're kept, and at what quality.
The finished file downloads right after export. Re-downloading the same file is free; its characters are already charged and not taken again.
Finished exports stay available for re-download for about 5 days (the periods are approximate and are refined as the service develops). If you need to keep a file for a long time, download it right after it's ready. Can't find your file — write to support.
Pay monthly for a plan or top up your character balance one-off. When paying yearly — a 15% discount.
| Plan | Price/mo | Characters per month |
|---|---|---|
| Free | 0 € | 10.000 |
| Starter | 9,00 € | 120.000 |
| Pro | 29,00 € | 500.000 |
| Business | 99,00 € | 2.000.000 |
| Scale | 299,00 € | 8.000.000 |
Current prices, limits and character packs are on the “Plans” page. Payment by Visa/Mastercard on the provider's secure page.
You're charged for the result — for the created file, not for previewing.
How to switch to another plan and how renewal works.
Open the “Plan & limits” section in the studio and pay for the plan you want — new limits and features turn on right after payment.
If you need a lower plan, choose it in the same section at the next payment. For questions about switching mid-period and recalculation — write to support, we'll help pick an option.
The plan sets: how many characters are included per month, the maximum characters per request, access to premium effects and smart text preparation, transcription minutes and API access. Free — for trials; the API and advanced features are available from the Starter plan and up. The full comparison table is on the “Plans” page. Custom voices are unlimited on all paid plans — creating a voice costs 10.000 characters from your balance; storing and using it is free.
Buying a plan is a subscription: the payment renews automatically (monthly or yearly) until you cancel. You can cancel at any time — in your account under “Plan & limits” — the “Cancel subscription” link or via the link in the payment confirmation; access remains until the end of the paid period. One-off character packs do not renew.
Character balance and plan are different things: changing the plan does not reset characters already bought on your balance.What you can pay with and what to do for organizations.
Payment goes through the provider's secure page: Visa/Mastercard bank cards. A receipt arrives at your e-mail after payment.
Online payment works by card. If an organization needs a contract or payment by invoice, write to support — we'll discuss the arrangement. The company details are published on the “Company details” page.
The account is designed for a single user — separate team seats and shared accounts are not available yet.Speech generation and transcription from code via the REST API.
X-API-Key header.A full description of endpoints, parameters and response codes is in the API documentation.
Email verification, password, changing details and deleting the account.
To voice text you need to verify your email. In the “Account” section request a code — it arrives at your email and is valid for 15 minutes. No email — check the “Spam” folder and the address, and request the code again.
On the sign-in page click “Forgot password?” and enter your email — a code for changing the password will arrive (valid 15 minutes). Didn't arrive — check “Spam” or write to support.
In your account, in the “Account” section, you can change the password — the current password is required to confirm.
You can't change the email address yourself. If you need to change the linked email — write to support from the current address.
In your account, in the “Account” section, there is account deletion — you'll need to enter the password to confirm. The account, voices, projects and history are deleted, and the session ends. The action is irreversible.
The account is designed for a single user — team seats and shared accounts are not available yet.Premium effects and smart text preparation are available on a paid plan; on the free one they are ignored. Check your plan or use the basic features.
Check the sound and playback permission in the browser, and refresh the page. If it didn't help — write to support and specify the browser and device.
The voice may have been deleted. If you didn't delete it — write to support, we'll help restore it.
Long texts are processed in the background. If a task is stuck for a long time — start the export again (characters for the failed attempt are refunded). If it repeats — write to support.
The engine is waking up or working through the queue. Retry the request in a few seconds.
You need to verify your email to voice text: the verification code is in the “Account” section.Most often the sample transcript doesn't match the recording word-for-word. Re-record the sample and write exactly what is said. A short, precise sample of clean speech helps.
Recreate the voice from a clean, even 30–90 second sample. If it repeats — write to support.
The Free plan includes 1 test voice; on paid plans the number of your own voices is unlimited (creating one costs 10.000 characters from your balance; storing and using it is free). If you see this message on Free — delete the test voice or upgrade to a paid plan.
You may upload only your own voice or a voice you have rights to (see “Voice rights and privacy”).Payment confirmation sometimes arrives with a delay — wait a couple of minutes and refresh the page. If the characters still didn't arrive — write to support and specify the e-mail, date, time and amount of the payment; we'll check and credit them.
Write to support with the payment details — we'll check and return the excess.
Try another card, check the card's limits and the confirmation from the bank. On a failed payment no money is charged.
A refund is processed by the same method the payment was made, and usually takes up to 5 business days (the periods are approximate and are refined as the service develops). For a refund request — write to support.
A plan subscription renews automatically. You can cancel it at any time in your account under “Plan & limits” — the “Cancel subscription” link — access remains until the end of the current billing period and no further charge is made.
Creating your own voice is a sensitive operation. Here are the rules and documents.
Free: on registration you get 10.000 characters, with no card. After that — a plan or a character pack.
It's a neural network, it reads with life. You can pin a version you like, and it will stop changing.
10 core languages with a quality guarantee: German, English, French, Spanish, Italian, Portuguese, Russian, Chinese, Japanese and Korean — plus 14 more languages in beta via auto mode (all languages). Each voice has its own native language.
Yes. A long text is exported in the background, and an email arrives when it's ready.
Yes: the voice is private, not published without moderation, and can be deleted. You may upload only your own voice or one with permission.
Previewing is free but limited to 3.000 characters at a time. To voice the whole text — click “Export”.
Top up your balance or upgrade the plan in the “Plan & limits” section. Remember: balance and plan are different (see How characters are counted).
Turn on “Make cleaner” in the “Sound” block, or recreate the voice from a cleaner sample.
Write to support from the “Contacts” page — we'll help.