Starting this month, Fujitsu and Fujitsu Social Science Laboratory, are offering “LiveTalk“, a new speech-to-text transcribing tool, to businesses, colleges and schools in Japan.
Knowing that meeting and educational settings can challenge people with hearing issues, Fujitsu is seeking to visualise everyday speech as text to make interactions more natural.
LiveTalk’s Universal Design
Everyone can use LiveTalk regardless of hearing ability, since the software:
- Detects and converts speech from multiple speakers to text on multiple screens.
- Enables real-time sharing of information even with no human transcriber present.
- Provides two-way communication with options for text input and PC stamp tools.
- Supplies a transcript for meeting notes and storage of conversation logs.
LiveTalk works by using headset and handheld microphones to transcribe spoken dialogue in near real-time to multiple devices via wLAN connections and with emoticon support.
Contributing To Social Good
Fujitsu has a clear social mission in building LiveTalk as a universal solution for students and employees to participate in everyday environments, regardless of hearing ability.
As a next move, Fujitsu aims to team with third parties to build LiveTalk into applications for emergency services and specialised situations such as healthcare or judiciary systems.
Japan led early in speech to text tools, piloting iPhone-based classroom captions for high school students in 2010, and captioning at tertiary level reviewed in 2012. Telecom firm NTT DoCoMo and game firm Nintendo also tested classroom captions (stored in the cloud) and two-way text to speech tools with output to a whiteboard or DSi device.