AI avatar video guide
Make a talking avatar from text to speech.
For creators, course makers and small brands who want a presenter without filming. Type a script, pick a voice, get a lip-synced avatar.
How it works
Text to speech avatar in four steps.
From a written script to a lip-synced talking avatar video, in the order you do it in Frameo.
Write or paste your script
Type your lines or paste them in. Keep sentences short so the voice pauses where a person would.
Choose the avatar
Upload a photo or describe a presenter. Front-facing, evenly lit photos sync best.
Pick a voice, language and pace
Choose a text-to-speech voice, set the language and slow or quicken the pace. Hear it before you commit.
Generate and export
Frameo lip-syncs the avatar to the voice. Export the finished video in 9:16 or 16:9.
Text to speech voices
Choose the voice your avatar speaks with.
Same line, five text to speech voices. Pick the one that fits your presenter.
Warm narrator
Unhurried and friendly. Suits tutorials and onboarding.
“Hi, I'm here to walk you through it in sixty seconds.”
Use this voiceUpbeat creator
Bright and quick. Suits social clips and product intros.
“Hi, I'm here to walk you through it in sixty seconds.”
Use this voiceCalm teacher
Even and patient. Suits lessons and course welcomes.
“Hi, I'm here to walk you through it in sixty seconds.”
Use this voiceCrisp corporate
Clear and measured. Suits team updates and explainers.
“Hi, I'm here to walk you through it in sixty seconds.”
Use this voiceSpanish voice
Natural Spanish. Suits welcome messages for Spanish speakers.
“Hola, estoy aquí para guiarte en sesenta segundos.”
Use this voice
Real output
Talking avatar videos and the scripts behind them.
›Medium wide shot, locked-off, of a friendly woman in her thirties in a teal shirt standing beside a blank whiteboard in a bright training room, speaking to camera and gesturing once toward the board. Soft even studio light from camera-left with bounced fill, neutral colour, no text, photoreal, 6-second 3:4 clip.
›Slow steady push in toward a friendly teacher in his forties in a mustard cardigan, seated at a wooden table in a cosy classroom corner, speaking warmly to camera with a gentle nod. Soft creamy window light, ivory tones, lifted blacks, matte film look, shallow depth of field, no text, 6-second 3:4 clip.
›Handheld shot with natural sway of a trainer in his thirties in a green sweater standing at a wooden workbench with neat tool rolls, answering to camera with a confident nod and a small shrug. Warm natural window light from the side at 3800K grazing wood and wool, lived-in mood, no text, 6-second 3:4 clip.
›Close-up of a warm woman in her forties in a cream sweater speaking to camera with a soft smile, lips moving clearly, a slight head tilt. Soft frontal amber lamp light, creamy warm-white grade, low contrast, dark subdued background, cosy evening mood, photoreal, no text, 6-second 3:4 clip.
Script to speech
Watch your script become speech.
Hi, welcome to the studio.
Today I will show you the basics.
Type a line. Hear it spoken.
Short lines give natural pauses.
Then export and share.
- Voice: warm narrator
- Pace: relaxed
- Language: English
Talking avatar text to speech questions
A talking avatar is an AI presenter: it reads your script in a text-to-speech voice with lip-synced mouth movement. You write the words, pick the voice, and Frameo makes the video.
Yes. Upload a photo and the avatar speaks your script. Front-facing, evenly lit photos give the cleanest lip-sync. You can also describe a presenter instead.
Frameo makes one video per run, so split a long script into scenes and keep each one short. For videos over five minutes, a longform run suits better.
Each voice shows its language and accent before you choose it, so listen first and pick the one that fits your presenter.
Yes. Upload your recording and the avatar lip-syncs to it, so the video carries your voice rather than a synthetic one.
Your first script
Type the first line your avatar will say.
Write it like you would say it. Frameo opens with your script filled in, ready for a presenter and a voice.