AI text to speech
Paste a script, choose a voice, say how it should sound, and get a voiceover you can play here and download. The text is read exactly as you wrote it.
- 12 voices
- About 130 languages
- Delivery directed in words
- 5 credits per 1K characters
Made with MakeMidas
Real results from the models on this site, with the prompts that made them.
It's almost midnight, and the city finally exhales. Somewhere a kettle clicks off, a train hums across the river, and the last bakery on Fifth Street pulls tomorrow's bread from the oven. Stay a while. The night is long, and the music is just getting started.
Script
It's almost midnight, and the city finally exhales. Somewhere a kettle clicks off, a train hums across the river, and the last bakery on Fifth Street pulls tomorrow's bread from the oven. Stay a while. The night is long, and the music is just getting started.
Voice: Kore · Delivery: “warm and unhurried, like a late-night radio host”
清晨六点,老城区的第一家早点铺掀开了蒸笼。白雾一下子涌上屋檐,豆浆的香味顺着巷子飘出去,叫醒了还在打盹的猫。老板娘一边擦着桌子,一边跟熟客打招呼:今天的包子,还是两个肉的一个菜的?
Script
清晨六点,老城区的第一家早点铺掀开了蒸笼。白雾一下子涌上屋檐,豆浆的香味顺着巷子飘出去,叫醒了还在打盹的猫。老板娘一边擦着桌子,一边跟熟客打招呼:今天的包子,还是两个肉的一个菜的?
Voice: Kore · Delivery: “亲切自然,像纪录片旁白”
Popular models
Script
It's almost midnight, and the city finally exhales. Somewhere a kettle clicks off, a train hums across the river, and the last bakery on Fifth Street pulls tomorrow's bread from the oven. Stay a while. The night is long, and the music is just getting started.
Voice: Kore · Delivery: “warm and unhurried, like a late-night radio host”
The most expressive, high-fidelity reading, in about 130 languages; direct the delivery in words.
8 credits per 1,000 characters
Script
清晨六点,老城区的第一家早点铺掀开了蒸笼。白雾一下子涌上屋檐,豆浆的香味顺着巷子飘出去,叫醒了还在打盹的猫。老板娘一边擦着桌子,一边跟熟客打招呼:今天的包子,还是两个肉的一个菜的?
Voice: Kore · Delivery: “亲切自然,像纪录片旁白”
Faster and cheaper, in about 101 languages; also directed in words.
5 credits per 1,000 characters
Narration for video, ads and lessons
Read a product demo, a course module, a podcast intro or an explainer. Each voice keeps the same sound from one script to the next, so a series stays consistent.
Voices with distinct characters
Twelve Gemini voices, six female and six male, from a firm ad read and a bright teacher to a friendly storyteller and a gravelly narrator. Each has a short note in the menu to help you pick.
Direct the delivery
Describe how it should sound in a few words, such as “warm and unhurried” or “excited sports commentator”, and the voice follows its tone, emotion and pace. Leave it empty for a natural read.
Spoken as written
Your text is never rewritten. Commas, full stops and question marks shape the pauses and intonation, so punctuate the script the way it should sound.
Keep creating
One account, one credit balance for images, video and audio.
Choose your plan
A monthly plan or a one-time credit pack gives every model and clean output.
Black Friday: 1,250 credits
- Base credits: 1,000
- Black Friday bonus: 250
- Valid for 12 months
Black Friday: 6,250 credits
- Base credits: 5,250
- Black Friday bonus: 1,000
- Valid for 12 months
Black Friday: 12,500 credits
- Base credits: 11,000
- Black Friday bonus: 1,500
- Valid for 12 months
Black Friday: 31,250 credits
- Base credits: 28,750
- Black Friday bonus: 2,500
- Valid for 12 months
Subscribers get 62,500
- Base credits: 62,500
- Valid for 12 months
Any pack unlocks every model and several outputs at once, even without a subscription.
Questions
Which model should I use?
Gemini 3.8 Flash TTS gives the most expressive, high-fidelity reading and suits narration, ads and audiobooks. Flash-Lite TTS costs less and returns faster, which suits drafts, app prompts and long batches. Both speak with the same voices.
How long can a script be?
Up to 5,000 characters per voiceover, about five minutes of speech at normal pace. Split longer texts into parts and keep the same voice and delivery.
How much does a voiceover cost?
It depends on the length of the text, per 1,000 characters (Gemini 3.8 Flash TTS: 8 credits; Gemini 3.8 Flash-Lite TTS: 5 credits), rounded up, with at least 1 credit per voiceover. The price updates as you type. If a voiceover fails, the credits are refunded automatically.
Which languages can it read?
Flash TTS reads about 130 languages and Flash-Lite TTS about 101, including English, Chinese, Japanese, Korean, Spanish, French, German, Portuguese, Hindi and Arabic. The language is detected from the text, so there is nothing to set.
Can I use the audio commercially?
The license of the model you used decides; MakeMidas adds no limits of its own. Check the model's terms before releasing a track on streaming services or in a paid product.