Romi is developed by mixi, Inc. one of Japan’s largest IT companies, and is an autonomous conversational robot that uses its AI (artificial intelligence) to think of words to say in response to the content and flow of the conversation.

Romi has a unique voice synthesizer created using the “Custom Voice” service. ReadSpeaker has developed a bespoke text-to-speech engine using voice actor Yoko Honna’s recorded voice. This technology allows Romi to interact in a natural voice that expresses the character of the robot. 

Find more information on Romi here.

Custom Voice

The Custom Voice TTS (text to speech) software service creates a synthetic version of a narrator’s voice. It is created using deep learning technology together with the customer’s preferred narrator. This technology also supports emotional speech synthesis to express happiness, sadness, and anger.

Find out more about Custom Voices from the ReadSpeaker VoiceLab.

Spoken Dialogue System with ReadSpeaker speechEngine SDK Embedded

ReadSpeaker speechEngine SDK Embedded is a stand-alone text-to-speech SDK (Software Development Kit) suitable for development on smartphones and other embedded devices. Since ReadSpeaker speechEngine SDK Embedded can create speech from text data in real-time, various automatically generated text data can be instantaneously transmitted to the user as voice.

A simple set of APIs in ReadSpeaker speechEngine SDK Embedded enables efficient and speedy development. It supports not only English, but also more than 20 languages.

Find out more about ReadSpeaker text-to-speech software here or contact us if you’d like to discuss an embedded development project.