“Play 3.0 mini'' can read text in male and female voices in more than 30 languages ​​including Japanese – GIGAZINE

It supports more than 30 languages ​​and also has a voice that can read sentences in multiple voices and accents.Play 3.0 MiniHe showed up. It also supports Japanese, and the selling point is the “nature of the sound.”

Introducing Play 3.0 Mini – a lightweight, reliable and cost-effective multilingual text-to-speech model
https://play.ht/news/introducing-play-3-0-mini/

You can try Play 3.0 mini for free by registering your email address or Google account from the page below.

PlayAI is a text-to-speech playground
https://play.ht/playground/

I want to try it in Japanese, so when I access it, I change “LANGUAGE” to “JAPANESE”.

Choose your favorite voice from “VOICE”.

Enter text in the input box and click “Create Speech” to create speech.

I tried reading the text out loud in several voices. Although he reads with a natural accent, I'm concerned that the audio is broken and some words aren't pronounced correctly. The average wait time for generation speed is 189 milliseconds.

I tried the “Play 3.0 mini” which can read sentences in a male/female voice – YouTube

Play 3.0 mini supports more than 30 languages ​​including Japanese, Hindi, Arabic, Spanish, Italian, German and French, in addition to the beta sites mentioned above.APIAvailable through. By creating applications using application programming interfaces (APIs), you can create conversation forms that respond to user feedback.

Introducing Play 3.0 mini – a new built-in text-to-speech model for real-time voice AI – YouTube

Please note that some use of the API is free, but mainlyexpensesit takes.

According to developer PlayHT, “hallucinations” can be seen in large-scale language models, just as in large-scale conversational language models, and most text-to-speech models are vulnerable to misreading alphanumeric characters. Play 3.0 mini has corrected this point and is compatible with use cases where there may be a problem if important information such as phone numbers, passport numbers, dates, etc. are read incorrectly.

PlayHT says, “The Play3.0mini is the fastest, most conversational voice model, with dramatically improved reliability and sound quality. An efficient, multilingual, AI-powered voice synthesis model that we plan to release in the coming months.” “Our goal is to make the model so small and cost-effective that it can Run it on users' devices.