Ondoku | Listen to 25 Free Next-Gen Japanese AI Voice Samples

Thank you for always using Ondoku.

We’re excited to announce the release of “Ondoku Beta,” a new next-generation AI voice that can read aloud “more naturally” and “more like a human”!

By adopting state-of-the-art speech synthesis AI, the new voices offer:

• Richer emotional expression
• Freely specify voice styles such as “cheerful” and “narration”
• All voices support English and multilingual speech synthesis

This makes it even easier to create high-quality spoken audio.

This article takes an in-depth look at new voice samples and how to use the latest features of the next-generation AI voices.

[Free] Introducing the next-generation AI voice “Ondoku Beta”!

Ondoku’s new AI text-to-speech voice, “Ondoku Beta,” is available here.

Why not try the next-generation AI text-to-speech voice, which can deliver rich emotional expression in Japanese and English?

AI Speech Synthesis [Beta] - Next-Generation TTS with Rich Emotional Expression | Ondoku

Generate expressive speech with the latest AI technology. Next-generation text-to-speech with 25 voices to choose from. Create natural-sounding speech by adjusting speed and pitch and specifying a reading style.

Types of next-generation text-to-speech voices (speakers) (Japanese, English, and multilingual support)

Ondoku’s new AI text-to-speech voices include female and male voices.

• Female: 12 voices
• Male: 13 voices

Every voice (speaker) has its own reading style and can read aloud with rich emotional expression in a voice that sounds just like a human.

What’s more, all voice types support multilingual speech synthesis!

You can generate speech in languages from around the world, including Japanese and English.

Listen to samples of Ondoku’s next-generation AI voices

Let’s take a look at some voice samples of Ondoku’s new AI voices.

The sample text read aloud is:

Hello, this is Ondoku.
Our voice is now more natural and more human-like!

Here’s an example in Japanese.

The audio was generated with the following settings:

• Speed: Normal (1x)
• Pitch: Normal (0)
• Reading style: Cheerful and energetic

Give it a listen—the speech sounds more natural and human-like than ever!

New AI voice samples (25 voices)

Japanese example read aloud by 12 female voices

Japanese example read aloud by 13 male voices

[Emotional expression] Next-generation AI voice settings (reading style, quality, speed, and pitch)

Ondoku’s new AI voices let you freely adjust the spoken audio to suit your preferences and needs!

You can configure and adjust the following three settings:

• Reading style
• Speed
• Pitch

Here’s an explanation of each setting.

Reading style | You can now freely specify styles and emotional expression!

The biggest update to the newly released next-generation AI voices is the ability to freely specify the reading style!

You can set the reading style in this section of the new voice text-to-speech page.

There are two ways to specify a style: pre-registered options and a free-text box.

How to specify a style using the pre-registered options

First, here’s how to choose a reading style from the pre-registered options.

Six options are available in the dropdown menu.

• News-style
• Narration
• Cheerful and energetic
• Calm
• Business
• Storytelling

If you want to quickly generate audio for video production, language-learning materials, or in-store announcements, simply choose a reading style from these options.

Voice samples by reading style (Japanese)

The voice used is “Ellis.”

Voice sample read aloud in “News-style”

Voice sample read aloud in “Narration” style

Voice sample read aloud in “Cheerful and energetic” style

Voice sample read aloud in “Calm” style

Voice sample read aloud in “Business” style

Voice sample read aloud in “Storytelling” style

How to specify a style and emotional expression by freely describing how the text should be read

There’s more!

You can now also freely describe the reading style to match the purpose and content of your audio.

To specify a reading style, enter your instructions in the “Reading style (optional)” text box.

For example, you can generate speech with rich emotional expression like this.

All samples use the voice “Ellis.”

Voice sample with the instruction “Please read this in a whisper.”

Hello, this is Ondoku. Our voice is now more natural and more human-like!

Voice sample with the instruction “Please read this in the voice of an 80-year-old.”

Hello, this is Ondoku. Our voice is now more natural and more human-like.

Voice sample with the instruction “Please read this in the Kansai dialect spoken in Osaka.”

Hello, this is Ondoku. It sounds more natural and more human-like now!

This makes it possible to read aloud in a far wider range of styles than with previous text-to-speech voices!

Why not try using the new AI voices to create speech just the way you want?

This article introduces detailed examples of how to use reading styles.

It explains more practical uses, such as emotional expression, age, and character voices, so be sure to check it out!

[Voice tuning] A complete guide to reading styles for the new AI voices. Create more human-like emotional expression and performances for free! | Ondoku text-to-speech software

A complete guide to voice-tuning techniques for the next-generation AI voice “Ondoku Beta.” Freely create emotional expressions, specify ages, character voices, dialects, and even foreign languages such as English. Includes samples and tips on how to tune AI text-to-speech voices that are free for commercial use.

Adjusting reading speed and pitch (voice height)

On Ondoku’s new text-to-speech page, you can also adjust the reading speed and pitch (voice height).

Reading speed can be adjusted across the following five levels:

• Very slow (0.5x)
• Somewhat slow (0.75x)
• Normal (1x)
• Somewhat fast (1.25x)
• Very fast (1.5x)

You can use these settings in many ways—for example, slower audio for language learning and shadowing, or faster audio for entertainment videos.

Pitch (voice height) can also be adjusted across the following five levels:

• Very low (-1)
• Somewhat low (-0.5)
• Normal (0)
• Somewhat high (0.5)
• Very high (1)

*The reading speed and pitch adjustment features are currently being refined. The adjustment range and settings may change in the future.

Ondoku's New AI Voices Support Multiple Languages! Read Aloud in Multiple Languages

Ondoku's new AI voices can read aloud in languages other than Japanese, too!

All voices support multiple languages, including English and other foreign languages!

For example, here's what it sounds like when reading aloud in English, French, Spanish, Korean, and Chinese.

All the samples are read aloud using the voice "Ellis."

English
Instruction: 「Please read aloud brightly and energetically.」

Hello, I'm Ondoku.
The voice has become more natural and more human-like!

French
Instruction: 「Veuillez lire à voix haute avec enthousiasme et énergie.」

Bonjour, je suis Ondoku. La voix est devenue plus naturelle et plus humaine !

Spanish
Instruction: 「Por favor, lee en voz alta con alegría y energía.」

Hola, soy Ondoku. ¡La voz se ha vuelto más natural y más humana!

Korean
Instruction: 「밝고 힘차게 읽어주세요.」

안녕하세요, Ondoku입니다. 더 자연스럽고, 더 인간다운 목소리가 되었습니다!

Chinese
Instruction: 「请明快有活力地朗读。」

你好，我是Ondoku。声音变得更自然、更人性化了！

Things to Keep in Mind When Reading Aloud in Multiple Languages with Next-Generation AI Voices

When reading aloud in English or other languages with the new AI voices, you need to specify the reading style in the language you want to use for reading aloud.

In these samples, too, the reading style is specified in each respective language.

If you give the instruction in Japanese, "Please read aloud brightly and energetically," the text will be read aloud with katakana pronunciation.

Also Supports Reading Aloud in Multiple Languages!

Ondoku's new voices also support reading complex text that uses multiple languages, which previously required SSML tags to specify with Ondoku's multilingual voices!

For example, for this sample sentence, specify the languages in the reading style field, and the text will be read aloud with pronunciation appropriate to each language, such as English or French.

The voice used is "Ellis."

Instruction: 「明るく元気に読み上げてください。英語の部分は英語発音で、フランス語部分はフランス語発音で、スペイン語部分はスペイン語発音で、ドイツ語部分はドイツ語のネイティブスピーカー発音で読んでください」

ありがとうは、英語で Thank you 、フランス語で Merci 、スペイン語で Gracias 、ドイツ語で Danke といいます。

※For instructions like this, it's also fine to write the reading style in Japanese.

Frequently Asked Questions About Ondoku's Next-Generation AI Reading Voices

Finally, here are answers to common questions about using the new reading voice "Ondoku Beta."

Q. How many characters can be read aloud?

A. You can read up to 400 characters at a time for free.

With a paid plan, you can read up to 1,000 characters at a time.

With a paid plan, you can increase the limit to up to 3,000 characters by enabling "Expand input to 3,000 characters" below the Ondoku Beta input field or "Character limit expansion" on the settings screen.

Q. How are the characters read aloud counted?

A. They are counted the same way as before. The character count is combined with the number of characters read aloud using the previous voices.

Q. Can I check previous reading results?

A. As with the previous voices, you can check them on the "History" page.

Q. Can I use it for free?

A. Yes, you can. As with the previous voices, attribution is required when using it for free.

For details on how to provide attribution, see this page.

Take a look!

How to Provide Attribution for Ondoku: Examples and Important Notes | Text-to-Speech Software Ondoku

We welcome you to use Ondoku for free. However, attribution is required when using it for free. We've also received questions such as, "So, specifically, how should I provide attribution?" This article explains how to provide attribution when using Ondoku for free.

Q. Can I use it for commercial purposes (business use)?

A. As with the previous voices, commercial use (business use) is permitted.

For details on free and commercial use (business use), please see this page.

What You Can Do with Ondoku: Commercial Use (Business Use) and Prohibited Activities | Text-to-Speech Software Ondoku

Ondoku can be used for commercial purposes (business use). Regardless of whether you are an individual or a business, using it with the aim of obtaining direct or indirect monetary or other benefits is considered commercial use. However, please note that Ondoku has prohibited activities. This article explains what you can and cannot do with Ondoku.

Try Using Ondoku's New AI Voices!

As you can see, Ondoku's new AI voices make it easy to generate high-quality audio that's even easier to understand than before!

Why not try Ondoku's AI voices for yourself? They're more natural and more human-like than ever.

AI Voice Synthesis [Beta] - Next-Generation TTS with Rich Emotional Expression | Ondoku

Generate expressive voices with the latest AI technology. Next-generation text-to-speech with 25 voices to choose from. Create natural-sounding audio by adjusting speed and pitch and specifying a reading style.