Introducing the DORODORA TTS Playground: AI Speech Synthesis Right in Your Browser

Reading donation messages aloud with an AI voice has become hard to leave out of stream content these days. When a viewer's message plays as speech, it changes the mood of the broadcast and creates a new kind of interaction between streamers and their audience.

DORODORA also provides AI TTS and Voice Cloning TTS along with its donation pages. Still, before applying a voice to a real broadcast, you often want to hear what it sounds like first. That is why DORODORA runs the TTS Playground, where you can try TTS right in your browser without installing anything or signing in.

This article covers the features of the TTS Playground available at dorodora.com/playground, how to use it, and how it connects to the DORODORA donation service.

Try the TTS Playground Now

What Is the TTS Playground?

The TTS Playground is a free TTS playground run separately by DORODORA. The TTS provided here relies only on local TTS models that run in your browser, not remote cloud models. You type a sentence, pick a model and voice, then play the result right away or save it as a WAV file.

No sign-up or payment is required. Just open dorodora.com/playground and start using it.

Key Features

The TTS Playground packs a few core features into a simple interface.

Browser-Based Local TTS

Everything runs inside your browser. Once you download a model file, it is stored in your browser so you do not have to download it again. Your text and generated audio are never sent to a server, which makes it comfortable for quick tests or previewing personal sentences.

Multiple TTS Models

You can choose from four local models in the TTS Playground.

Model	Size	Voices	Languages
SuperTonic	399MB	10	31+
Pocket	199MB	8	English, German, Italian, Portuguese, Spanish
Kokoro	92MB	28	English
Kitten	28MB	8	English

SuperTonic supports more than 31 languages including Korean. You can pick a language from the list and combine it with any voice, which is handy for broadcasts with a global audience.

Pocket is a lightweight model split by language. Choose one of English, German, Italian, Portuguese, or Spanish, and thanks to the voice cloning feature you can also upload a voice sample of about 1-3 seconds and generate TTS that speaks exactly like the uploaded voice.

Kokoro and Kitten are English-only models. They are the smallest ones, so downloads are quick and they are great for casually testing English speech synthesis.

Animalese Effect

Animalese is not a model but an effect applied to the generated audio. Turn the toggle on and the output changes into a fast, cute character voice like an animal villager from a certain game. It works well for playful streams or short sound effects.

Up to 1,000 Characters

You can enter up to 1,000 characters at once. That is long enough to preview everything from short donation messages to full paragraphs.

WAV Download

Click the download button next to the player to save the generated audio as a WAV file. The model name is included in the file name, so files stay easy to tell apart. It is convenient when you want to use pre-made audio as stream material or process it further in an audio editor.

How to Use It

The TTS Playground works in a few simple steps.

Open dorodora.com/playground.
Pick a model at the top. On first selection the button shows the model size, and pressing it downloads the files.
Type the sentence you want to convert.
Choose a language. Available languages depend on the model.
Pick a voice.
Turn on the Animalese toggle if you want.
Press the Generate TTS button.
Play it in the player or click download to save the WAV file.

Because the feel changes with each combination of model, language, and voice, trying several combinations is recommended.

Using a Custom Voice

When you select a Pocket model, you can choose Custom in the voice list. This option uses voice cloning to turn your own voice into TTS exactly as it is.

Follow these steps to use a custom voice.

Select one of the Pocket models.
Choose Custom in the voice list.
Upload a noise-free voice file of about 1-3 seconds.
Enter your text and generate.

Your sample is processed inside the browser, and the custom voice is stored there too, so you can load it again on your next visit. It is a good way to preview how it feels before having donation messages read in your own voice.

When Is It Useful?

You can already hear the AI TTS voices on the DORODORA main page and donation pages, but the TTS Playground is worth using beyond that. It is free, and the audio you create can be used commercially, so it shines whenever you need speech for purposes outside streaming.

When you need speech synthesis without any cost
When you need voices for non-streaming content such as video dubbing, narration, or announcements
When you want to save the result as a WAV file for editing or other projects
When you want to check how a clone of your own voice sounds
When you need speech in more than 31 languages including Korean

Since all it takes is a browser, with no account or payment method, it is a tool you can casually reach for whenever you need a voice.

How It Relates to the DORODORA Donation Service

The TTS Playground is a tool for previewing part of the DORODORA donation service ahead of time. On a real donation page, messages from supporters are read by AI TTS, and alerts appear on stream through an OBS browser source.

Once you find a model and voice combination you like in the Playground, turn on TTS in the message settings of the DORODORA console to apply it to your donation alerts. Voice Cloning TTS is also provided, so just like the custom voices you upload in the Playground, supporters' messages can be read in your own voice. The Playground is a quick check before configuring, and the actual service flows through the donation page and stream console.

FAQ

Is the TTS Playground free?

Yes, it is a free tool you can use right away at dorodora.com/playground. You can test speech synthesis in your browser without signing up or paying anything.

Which browsers can I use?

It works in Chrome, Edge, Firefox, Safari, and other browsers that support modern features such as WebAssembly and the Web Audio API. Mobile browsers can open the page too, but a desktop environment is more stable considering the model downloads and execution.

Do I have to download the model every time?

No. Models downloaded once are stored in your browser and reused. Models that are already prepared show up as selected.

Are generated voices stored on a server?

No. Everything in the TTS Playground is processed inside your browser. Your typed text, uploaded audio, and generated voices are never sent to DORODORA servers.

Can I use it commercially?

Yes, audio created in the TTS Playground can be used commercially in streams, videos, and other content. The generated audio is saved as a WAV file, so you can use it right away for editing or publishing.

What audio file should I upload for a custom voice?

The Custom voice of a Pocket model works through voice cloning. A noise-free voice file of about 1-3 seconds is recommended, and samples with clear pronunciation produce more reliable results.

Can I use the Playground voices for real donation alerts as they are?

Turning on TTS in the DORODORA console makes donation messages read by an AI voice. Use the Playground to decide which model and voice you like, then connect it through the console settings so donation messages during your stream get a similar-sounding voice.

Wrap-Up

The DORODORA TTS Playground is a free tool for testing AI speech synthesis right in your browser, with no installation or sign-up. It supports the SuperTonic, Pocket, Kokoro, and Kitten models, and SuperTonic handles more than 31 languages.

You can add the Animalese effect to generated audio or save it as a WAV file, and all processing happens locally in your browser. Since it is free and allows commercial use, give it a try when picking a voice for donation alerts or when you need speech for your content.

You can try TTS right now at dorodora.com/playground.

Try the TTS Playground Now