4 voices · Tokyo standard · 48 kHz
Japanese text to speech
Four Japanese voices in standard Tokyo speech: a polite shop announcer, an elegant documentary narrator, a steady presenter, and a gentle grandfather who tells folk tales. Paste your script in ordinary Japanese, with kanji, hiragana and katakana mixed as you would normally write, and hear it read naturally.
Listen
Every Japanese voice
Akiko
Female · mature · Tokyo standard
“春の訪れとともに、古い城下町は淡い桜色に包まれます。川沿いの小道には、今年も多くの人々が足を運びました。”
With the arrival of spring, the old castle town is wrapped in pale cherry-blossom pink. Once again this year, many people came to walk the paths along the river.
- documentary
- narration
- elegant
Haruto
Male · adult
“こんにちは。本日はご案内いたします。”
Hello. I will be your guide today.
- japanese
- polite
- steady
Isamu
Male · elderly · Tokyo standard
“むかしむかし、ある山奥の村に、正直者のおじいさんとおばあさんが住んでおりました。ある日、おじいさんは山へ芝刈りに出かけました。”
Long, long ago, in a village deep in the mountains, there lived an honest old man and his wife. One day, the old man went up the mountain to cut brushwood.
- storytelling
- folk tales
- gentle
Yui
Female · young adult · Tokyo standard
“いらっしゃいませ。本日はご来店いただき、誠にありがとうございます。何かございましたら、お気軽にお声がけください。”
Welcome. Thank you very much for visiting our store today. If there is anything we can help you with, please don’t hesitate to ask.
- assistant
- announcer
- polite
About Japanese in Voxmint
Every Japanese voice in the catalogue speaks standard Tokyo Japanese, the variety used by national broadcasters and most corporate audio. Yui has the bright, courteous tone of an in-store announcement, Akiko narrates like a travel documentary, Haruto is a calm, even presenter, and Isamu tells mukashi-banashi, the old folk tales, the way a grandparent would at bedtime.
Write your script the normal way, in mixed kanji and kana. You do not need to add furigana or convert to hiragana. Most readings come out right from context, including polite forms like ございます and いたします that carry much of the tone in service announcements.
Two things are worth checking before you publish. First, names and rare kanji: a place name or surname with an unusual reading can be read the common way. If that happens, write that word in hiragana or katakana and it will follow your spelling. Second, pitch accent: Japanese distinguishes some words by pitch alone, and a synthetic voice occasionally puts the accent in the wrong place. It is uncommon on everyday vocabulary, and a fresh take or a slightly reworded sentence usually fixes it.
Punctuation drives the pacing. 、 gives a short breath and 。 a full pause, so a long sentence split with commas sounds far more natural than one unbroken line. Write numbers, times and dates as you want them spoken where the reading could vary: 午後三時 rather than 15:00, for example. For announcements with a fixed structure, keep sentences short and consistent, the way a real station or shop announcement is written.
Choosing between the voices is mostly about register. Yui and Haruto use polite desu and masu forms and keigo naturally, so they suit anything customer-facing. Akiko’s narration fits longer, descriptive writing, and Isamu is at his best with the old-fashioned storytelling phrases of folk tales. If you need a younger, casual voice for anime-style or game content, design one in the studio by describing its age and personality.
Design or clone a Japanese voice
Describe the voice you want, for example:
A calm Japanese woman in her late twenties with a clear, gentle voice and a polite, measured delivery, like an NHK news announcer reading the evening headlines.
Or clone one from a few seconds of your own recording.
Use cases
What people make with Japanese voices
Shop, station and phone announcements
Yui’s polite keigo delivery fits store announcements, queue systems and phone menus, where the tone has to be courteous without sounding stiff.
E-learning and language study
Clear standard pronunciation makes these voices useful for vocabulary drills, listening exercises and course videos for learners of Japanese.
Documentary and travel narration
Akiko’s unhurried, elegant narration suits travel films, museum guides and brand stories that need a calm, premium feel.
Picture books and folk tales
Isamu reads like a grandfather telling mukashi-banashi, which makes him a natural fit for children’s audio, bedtime apps and storytelling channels.
Games and interactive characters
Design a character voice in the studio, save it to your library, and reuse it for every line of dialogue so the character sounds the same from the first scene to the last.
Questions
- Does it read kanji correctly?
- Almost always for everyday vocabulary. For names or rare readings, write the word in kana and the voice will follow your spelling exactly.
- Can it speak regional accents like Kansai-ben?
- The ready-made voices all speak standard Tokyo Japanese. You can describe a Kansai speaker in the studio, but results vary, so listen to several takes before relying on one.
- Can I mix English words into a Japanese script?
- Loanwords written in katakana read naturally. For English words in Latin letters, writing them in katakana gives you more control over how they sound.
- How do I add pauses?
- Use 、 for a short pause and 。 for a longer one, and break long sentences into shorter ones. That usually matters more than any other setting.
- Can a character keep the same voice across a whole game or series?
- Yes. Once you have a voice you like, save it to your library and every new line uses the same timbre. Cloning from a clean reference clip, of your own voice or one you have permission to use, gives the most consistent result.
- Can I clone a Japanese voice?
- Yes. Upload or record a few seconds of clear Japanese speech in your own voice, or in a voice you have permission to use and new lines will be spoken in that timbre; adding the transcript of your clip gives the closest match.