Write the dialogue your material does not have.
Tell Kukarella what conversation you need: the language, the level, how many people speak. It writes the script, creates the speakers and gives each one a voice. Press play and you hear it. Then pick the accent your class needs. This takes about 30 seconds, and you do not need an account or a card to try it.
2,000+ voices in 152 languages and accents. Every one plays a sample before you sign up.
What dialogue shall we write?
Free, and no account needed to hear the result. Three requests without an account; a free account gives you ten.
Or start from one of these:
You can also say your request out loud, or upload your own document and we will turn it into a conversation. Prefer to write it yourself?
This is what came back
One sentence in. A finished script, the speakers made, and every line already given to one of them. No audio yet, and nothing spent.
Write a dialogue at a pharmacy for B1 Spanish learners, about eight exchanges long.
The speakers it made
Lucía
Nova · reads any language
Andrés
Echo · reads any language
Hear the draft
Eight short exchanges. Present tense. Words a B1 group already knows. Written to be heard, not read.
- LucíaBuenos días. ¿En qué puedo ayudarle?
- AndrésBuenos días. Llevo tres días con dolor de garganta.
- Lucía¿Tiene fiebre?
- AndrésUn poco, por las noches.
- LucíaLe puedo dar unas pastillas. Una cada ocho horas, después de comer.
- Andrés¿Necesito receta?
- LucíaPara estas no. Si en tres días sigue igual, vaya al médico.
- AndrésMuy bien. Muchas gracias.
Nothing is spent yet
No audio is made until you press play.
Every word is editable
The script is a document. Change it like one.
These two voices read any language
They are not Mexican voices. You choose that yourself — that is the next section.
Then you choose the voices
The assistant hands you voices that read any language. They are good enough to check the script. They are not Mexican voices, and the assistant will not guess that your class needs one. Choosing the accent is your step, and it takes seconds.
Fine for checking that the script is right. The Spanish will not sound Mexican, because these voices have one fixed accent.
seconds
Mia S. is open to you now, without an account. Gerardo needs a paid plan — you can hear his sample before you decide.
Change all the speakers, not one.
Voices from different families sound different next to each other. If you give one speaker a native Mexican voice and leave the other on a draft voice, the two will not match. Change them together.
Try it with your own sentence
Prefer to write it yourself?
Listen to every voice before you sign up
No account. No form. No email address. Every voice in the library plays a sample — including the ones a paid plan opens. Many people come here for one thing: the voice with the accent their class needs. Your students do not need “Spanish”; they need the Spanish they are going to hear.
Choose a language, then a country.
English (UK), English (Ireland), English (Singapore). Spanish (Mexico), Spanish (Chile), Spanish (Spain). This is the real list, not a sample.
Listen to any voice.
No account. No form. No email address. Every voice in the library has a sample you can play.
A lock means the voice is not open to you yet.
Hover it and it says what opens it — a free account or a paid plan. Either way you can still play its sample.
Each voice is labelled native or multilingual.
The label sits next to the name, so you never have to guess. Which kind you pick changes how your Spanish lesson sounds. The next section explains the difference.
Some accents are deeper than others.
Some countries have dozens of native voices and some have two. Open a country and you see exactly what is there, before you write anything.
Native or multilingual, and why some voices sound wrong
Choosing the wrong kind of voice is the most common reason people think an AI voice sounds bad. There are two kinds and they are good at different jobs. The voices the assistant hands you are multilingual — which is exactly why the accent is a step you take yourself.
Native voices
A native voice speaks one language only. It knows how that language really works: where the stress goes, how numbers are read, how a question sounds. It is clear and correct.
It has a real regional accent and you choose which one — Spanish from Mexico, Chile or Spain; English from Britain, Ireland or America. A native voice cannot read another language.
Multilingual voices
One multilingual voice reads many languages. This is useful when one character has to speak several languages.
None of them is excellent in any one language. After about fifteen seconds most listeners hear a mistake in stress or pronunciation. And each one reads a language with one fixed accent — usually the accent of its English recording, which is why Spanish can come out American-sounding.
So how do you choose?
One language, and the accent matters
Use a native voice. For a language lesson this is almost always the right answer.
One voice that must read several languages
Use a multilingual voice, and accept the accent it comes with.
If your Spanish sounds a little American, or your Finnish has an English t and r in it, you are using a multilingual voice. Change to a native voice and the problem goes away.
Your learners hear more than just you
You are one voice in the room. One accent. One speed. One way of saying things.
Your students learn to understand you very well. Then they meet somebody else, and they are lost.
Your course book has audio, but it was recorded for a different class and a different syllabus. The level is right for some of your students and wrong for the rest.
Kukarella gives you a group of speakers. You decide what they say.
Listening practice for your real class
The same situation at A2 for Tuesday and at B2 for Thursday. Each one takes about a minute to ask for.
The situations your students will really be in
A pharmacy. A job interview. A landlord who will not repair the heating. An angry customer.
A real conversation, not one person talking
People interrupt each other, disagree and misunderstand. That is what listening practice needs.
A voice that is not yours, when that works better
A computer voice can correct a mistake without the student feeling blamed.
What you can say when somebody asks why you bought it
I make listening material for our own syllabus, at any level I need, in the accents my students will really hear. And I do not have to buy another course book.
Make the first one for your class
Prefer to write it yourself?
Two more, made the same way
Each script below came from the one sentence printed above it. We picked the voices ourselves — that is the two-click step.
Write an IELTS Listening Part 1 about booking a coach trip, two speakers, British English.
IELTS Listening Part 1 — booking a coach trip
- AgentHighland Coaches, good afternoon.
- CallerHello. I would like to book two seats to Inverness on the fourteenth.
- AgentThe fourteenth of March. That is the 9:40 or the 14:15.
- CallerThe 9:40, please. Is there a student price?
- AgentYes. Twenty-two pounds fifty each, with a student card.
Names, dates and prices are placed so students have something to catch. This is how the exam does it.
Write a short role-play for A2 English learners ordering coffee, using only the present simple.
A2 English — ordering coffee
- BaristaHi there. What can I get you?
- CustomerA large coffee, please. With milk.
- BaristaAnything to eat?
- CustomerNo, thank you. How much is that?
- BaristaThree eighty.
Present simple only. Short enough for students to say back.
Ask for a level, in any language, and get that level
Name a level and you get a script at that level, not one that only mentions it. You do not have to name a level at all — say “for people who are not native speakers”, or “for a first-year class”, or “for somebody new to the subject”, and you get that instead. This works in the language you teach, not only in English. CEFR covers Spanish, French, German, Italian and the rest of Europe. HSK is Chinese, JLPT is Japanese, TOPIK is Korean.
Levels
- CEFR A1
- A2
- B1
- B2
- C1
- C2
- HSK 1–9
- JLPT N5–N1
- TOPIK 1–6
- “beginner”
- “elementary”
Exam formats, in the language the exam is taken in
IELTS
English
TOEFL iBT
English
TOEIC
English
DELE
Spanish
DELF
French
Goethe
German
HSK
Chinese
Any language, not only English
Ask in the language you teach and you get that language. Not the language of the file you uploaded. Not the language of your source text. If you asked for German, you get German.
No sentence is longer than 30 words
This holds at any level. Long sentences break the rhythm of a spoken voice, and a listening exercise made of them is one nobody can follow.
Ask for a level now
Name the level and the language. That is the whole request.
A taste of what you can ask for
Any language, any level, any length, any tone. Press one of these and it goes into the box above. Then change a word, or send it as it is.
These are only examples. You can type your own request at any time, in your own words.
A free account triples what you can try with
30,000 characters instead of 10,000, and 30 minutes of transcription instead of 10. Every voice in the library opens on any paid plan.
The AI writes a draft. The project is yours.
Nothing the assistant makes is locked. Every part of it is a thing you can change afterwards.
Change the voices
Swap any speaker and every line they have changes with them. Change them all together, so they match.
Add or remove speakers
Make a new one, give them a name, a photo and a voice. Save a group and use it again next time.
Edit the text
Edit it like a document. Or write the whole conversation yourself and never use the assistant.
Control how each line is spoken
Ten pause lengths, from half a second to a minute and a half. 2,900+ sound effects, and you can add your own audio. Speed, pitch and volume work on Microsoft and MiniMax voices, not on OpenAI’s.
Ask the AI to change one paragraph
Rewrite it, make it simpler, translate it. The voices stay where they are.
Download what you need
MP3 or WAV, one file or split by paragraph. The text exports as TXT or DOC.
Share it
Invite up to 10 people as editors or viewers. Or send a link that plays without an account.
Prefer to start from nothing?
Make an empty project, add your speakers and type your own script.
Money, and what happens if you stop
These are the four questions people ask us before they pay. The answers do not change.
If I stop paying, do I lose my work?
No. Your projects stay, and everything you already turned into audio stays with them — you can still play it and download it. Files you uploaded can be removed if you use more storage than your plan allows. A paid plan is
What if it is not what I expected?
Ask for your money back within 15 days of your first payment. Write to support@kukarella.com. Cancelling is different, and you can do it in your profile at any time. If you cancel and come back within 30 days, your unused credits are still there.
Can I use it in a course I sell?
Yes, on a paid plan. Commercial rights to every AI voice are included, as long as you hold the rights to the text you turn into speech. The free plan does not include them — it is for trying the product and learning how it works.
I only need this once.
Then take a plan for a single month — for one project that is almost always the cheaper route. Here is what each way of paying actually gives you.
Prime Lite, Prime or Prime Plus. That month’s credits, every voice in the library, and commercial rights for the audio you make. A plan does not stop by itself: if you need the one month only, cancel it before the month ends. You can do that yourself, at any time.
About a third cheaper than paying month by month, and every monthly allowance multiplied by twelve. This is the one to take if you will be making material all year.
For a different case: your plan is right, and you just need a few more credits this month rather than moving up a tier. Bought once, never expires. It does not change which voices you can use.
Questions people really ask
Taken from our own mailbox, not invented. The answers here are the answers we give.
Can I try it without an account?
Yes. Describe the conversation you want. The assistant writes it, creates the speakers and gives each one a voice. Press play and you hear it. A free account is needed to save what you make, and a paid plan opens the rest of the voice library.
Can it make dialogues in my language?
Almost certainly. Kukarella covers 152 languages and accents. Dutch, Turkish, Finnish, Vietnamese, Arabic, Japanese, Korean and Chinese are all there. Write your request in the language you want, or just name the language.
Which voices does the assistant choose?
It picks from the same fixed set of 22 voices, for everyone and on every plan. The language of your script does not change that set. If the accent matters, change the voices yourself afterwards. That takes seconds, and the whole library is open to browse first.
Why did my Spanish sound American?
You were using a multilingual voice. A multilingual voice reads a language with one fixed accent, and often with its English recording — which is why your Spanish came out American-sounding. Change the speaker to a native Spanish voice and choose the country you want.
Can I change the voices the AI chose?
Yes, all of them, at any time. You can listen to a voice on its own before you decide. When you change a speaker’s voice, every line that speaker has changes too. Change all the speakers together so the voices match.
We chose one voice for its accent and use it for all our audio. Will it still be there next year?
Every voice in the library is made by another company. If that company stops offering the voice, it can leave our library too. This is rare. If it happens, send us a few recordings you made with that voice. We can make a clone of it and add it to the library for everyone. The clone sounds very close to the original, and it will most likely read more than ten languages.
Can I write the dialogue myself?
Yes. Make an empty project, add your speakers, type or paste your text, and give each line to a speaker. The assistant is the fast way in, not the only way.
How long does it take?
About 30 seconds from typing your request to hearing the audio. Under a minute in normal use.
What level can it write at?
CEFR A1 to C2, HSK 1 to 9, JLPT N5 to N1, TOPIK 1 to 6, or plain words like “beginner”. For exam formats: IELTS, TOEFL iBT, TOEIC, DELE, DELF, Goethe and HSK, by part.
Can I use the audio in a course I sell?
Yes, on a paid plan. Commercial rights to all AI voices are included, as long as you hold the rights to the text you turn into speech. Not on the free plan.
Can I turn my own material into a conversation?
Yes. Upload up to 10 files — documents, articles, notes, even images — and ask for a conversation built from them.
Can I share a dialogue with my students?
Yes. Send a link that plays without an account. Or invite up to 10 people as editors or viewers if you are working together.
What can I download?
Audio as MP3 or WAV, as one file or split paragraph by paragraph. Split files are easier to use in a video. The text exports as TXT or DOC.
Do you have an API?
No. Kukarella has no public or developer API, and we are not planning one at the moment. Everything happens inside the app.
I only need this once. Do I have to subscribe?
Not for long. For one project the cheapest route is usually a single month of a plan: you get that month’s credits and every voice in the library. A plan does not stop by itself, so if you need one month only, cancel it before the month ends. You can do that yourself, at any time. A top-up is for something else — when your plan is right and you only need a few more credits this month. A top-up does not open the voice library.
What do I get for making a free account?
More to try with: 30,000 characters for turning text into audio instead of 10,000, and 30 minutes of transcription instead of 10. Projects you make after signing up are saved to your account.
Something is not working. Who do I ask?
Ask us directly. The chat is the message icon in the bottom-right corner of the site. Or write to support@kukarella.com.
What dialogue shall we write?
Free to try. No card needed. You can hear every voice in the library before you decide anything.
Already convinced?
Prefer to write it yourself?