How to make a video in your own voice with AI

To make a video in your own voice with AI, record about a minute of clear reading in a quiet room, let the tool make a custom voice from it, then pick that voice to narrate your script. In Chasca, you open My voices, read a short script on screen, and your voice is ready in about a minute. From then on, any video you make can be narrated in your voice, in the language you recorded it in. If you have already recorded the narration itself, you can skip the clone and let that recording narrate the video instead.

What is a custom AI voice?

A custom voice, often called a voice clone, is a voice model made from a recording of you. You read a short passage, the tool learns your pitch, accent and rhythm, and from then on it can read any new text in a voice that sounds like yours.

That changes what a script costs you. Without a clone, every new video, and every fix to an old one, means setting up the microphone and recording again. With one, you edit the words as text, and the narration follows.

Why narrate a video in your own voice?

Because people know your voice. A teacher’s students, a creator’s audience and a founder’s customers have already heard you, and a familiar voice carries some of that trust into the video.

Learning research used to make a stronger case. In 2003, Richard Mayer and colleagues found that students who heard a narrated animation in a human voice did better on a problem-solving test than students who heard a computer voice, with a large effect. In Mayer’s later review, this “voice principle” held in 5 of 6 experimental comparisons, with a median effect size of d = 0.74.

Synthetic voices have improved since then. When Scotty Craig and Noah Schroeder repeated the comparison with a modern text-to-speech voice, learners rated it as credible as a recorded human voice and learned at least as well from it, and a follow-up study found the modern voice and the human voice were not statistically different on most measures.

So a good synthetic voice no longer costs your viewers anything. The reason to use your own is identity: the video sounds like the person who made it.

How much audio do you need?

Less than most people expect. Instant voice cloning tools typically ask for anywhere from a few seconds to a few minutes of clean speech. Studio-grade clones trained for hours on a voice exist, but they are a different product with a different price.

What matters most is that the sample is clean:

• One speaker. No interviewer, no co-host, no audience.
• No music or background noise. A fan, a fridge or a busy street ends up in the clone.
• Your natural pace. Read the way you explain things, not the way you read a legal notice.

In Chasca, the recorder asks for one minute, needs at least 20 seconds, and stops by itself at one minute. A full minute of clear reading sounds most like you.

How do you record a good voice sample?

You don’t need a studio. A quiet room and a few habits get you most of the way:

1. Pick a small room with soft furnishings. Curtains, a rug and a sofa absorb echo. Bare walls and a tiled floor bounce it back into the microphone.
2. Use the best microphone you have. A USB microphone or a headset beats a laptop’s built-in one. Even a phone’s earbuds with a microphone usually beat the laptop.
3. Keep the same distance. About a hand’s width from the microphone, and don’t turn away while you read.
4. Do a test take. Listen back on headphones before you record the real one. If you hear the room, move.
5. Read like you explain. Smile a little, keep your energy up and let sentences end naturally. The clone copies your delivery along with your voice.

How do you make a video in your own voice with Chasca?

1. Open My voices in the app and choose Add a voice.
2. Pick the language this voice will speak.
3. Read the script on screen. It starts with a short line saying the voice is yours, which keeps your consent on the recording itself. The recorder stops at one minute.
4. Listen back, record again if you like, then send it. Your voice is ready in about a minute, with a one-minute sample to check.
5. Make a video. Type a topic or drop in a document, then pick your voice in the narration options, above the narrators. Chasca writes the script, draws each scene and narrates it in your voice.

Every plan includes custom voices at no extra cost: Free and Solo come with one, Plus with 3, Pro with 5, Max with 10, and Team and Enterprise with 30. The custom voices page has the details.

Can your voice speak other languages?

In Chasca, no: each custom voice speaks the language it was recorded in, so a video it narrates is in that language. If you teach or sell in two languages, record a voice in each. Some tools can make a cloned voice speak languages you don’t, but the accent and the pronunciation are hard to check if you don’t speak the language yourself.

If you need a lesson in many languages, Chasca’s 30 narrators each speak all 79 languages in the same voice, so a course can keep one narrator from English to Arabic.

What are the rules on voice cloning?

The short version: your own voice is yours to use, and anyone else’s voice needs their permission.

• Consent. Tennessee’s ELVIS Act, signed in March 2024, added a person’s voice, including a simulation of it, to the rights the state protects, and it reaches the people who publish an unauthorized voice and the tools made to produce one. Other places have their own publicity and privacy laws.
• Disclosure in the EU. Article 50 of the EU AI Act, which applies from 2 August 2026, requires providers of AI systems that generate synthetic audio to mark it in a machine-readable way, and people who publish AI-generated audio that falsely appears to be a real person to say so.
• Fraud. The US Federal Trade Commission ran a Voice Cloning Challenge in 2024 because cloned voices are used in scams. Never use a voice to impersonate anyone.

For your own explainer videos, a line in the description such as “narrated with an AI voice made from my own” keeps viewers informed and costs nothing.

When should you record the narration instead?

A custom voice is best for scripts you will write and edit as text: lessons, product explainers, updates. When you have already recorded something worth keeping word for word, such as a lecture, a talk or a professional voiceover, use the recording itself.

In Chasca, attach the recording on the Create page and tick Use as the narration. The video keeps every word as you said it, and the board draws in time with your voice, for up to 1 minute on Free and 30 minutes on paid plans. Our guide to turning an audio recording into a video covers that path.

A checklist for your voice sample

• A quiet, soft room.
• The best microphone you have, at a steady distance.
• One speaker, no music.
• About a minute of reading at your natural pace.
• A test take, checked on headphones.
• Only your own voice, or a voice you have permission to use.

Sources

• Mayer, R. E., Sobko, K., & Mautone, P. D. (2003). Social cues in multimedia learning: Role of speaker’s voice. Journal of Educational Psychology, 95(2), 419–425.
• Mayer, R. E. (2014). Principles based on social cues in multimedia learning: Personalization, voice, image, and embodiment principles. In The Cambridge Handbook of Multimedia Learning (2nd ed.). Cambridge University Press.
• Craig, S. D., & Schroeder, N. L. (2017). Reconsidering the voice effect when learning from a virtual human. Computers & Education, 114, 193–205.
• Craig, S. D., & Schroeder, N. L. (2019). Text-to-speech software and learning: Investigating the relevancy of the voice effect. Journal of Educational Computing Research, 57(6), 1534–1548.
• Office of the Governor of Tennessee (2024). Gov. Lee signs ELVIS Act into law.
• European Commission. Transparency obligations under Article 50 of the AI Act.
• US Federal Trade Commission (2024). Approaches to address AI-enabled voice cloning.