WITH ALS Launches Voice Synthesis Communication Service 'Project XrossVoice'

WITH ALS, a general incorporated association, will begin offering the cloud-based voice synthesis communication service 'Project XrossVoice' on June 16, 2026, utilizing cross-lingual voice synthesis technology developed by NTT Human Information Laboratory.

'Project XrossVoice' is designed for individuals who have lost their voice due to illness or other reasons, or who may lose it in the future. The service learns and reproduces a person's unique voice from minimal past voice data, enabling natural communication. This service emerged from 'Project Humanity,' a collaborative initiative between WITH ALS and NTT aimed at creating a well-being society where everyone can live authentically. The technological foundation leverages NTT's advanced voice AI technology.

Unlike traditional voice synthesis services that require extensive prior voice recordings, 'Project XrossVoice' can generate personalized original voices even from limited historical data. Moreover, it supports speech output in English, Chinese, French, Korean, and Spanish, based solely on Japanese voice data.

WITH ALS will serve as the customer-facing provider, offering the service via a subscription model.

The service will be publicly demonstrated at 'MOVE FES.2026,' held on June 20, 2026, at EX THEATER ROPPONGI, where CEO Masaki Mutō (EYE VDJ MASA) will use it live during both performance and talk show segments. During the live set, he will engage the audience with Japanese and English calls, while in the talk segment, it will serve as his primary communication tool.

About Project XrossVoice

'Project XrossVoice' is a cloud-based voice synthesis communication service that recreates a user's authentic voice from past recordings and enables speech through text input. Leveraging NTT's voice AI technology, WITH ALS acts as the service provider and support interface, designed with deep empathy for ALS patients.

The interface has been co-developed through extensive testing with ALS users, aiming for universal usability. It supports diverse communication scenarios—from daily conversations and work meetings to presentations and public speaking events.

Masaki Mutō, CEO of WITH ALS, already uses the service in daily life, particularly valuing its utility in English-language meetings and public appearances.

Development Background

Conditions like ALS (Amyotrophic Lateral Sclerosis) can progressively impair speech while preserving physical function. Voice is not merely a tool for conversation—it embodies identity, memory, and social connection. Traditional voice banking requires lengthy recordings before speech deteriorates, posing physical and time-related barriers. Many patients cannot collect sufficient data after diagnosis or disease progression. WITH ALS launched this service to empower individuals to continue expressing themselves authentically through technology.

Key Features of the Service

1. Recreation of Personal Voice from Minimal Past Data

Even users unable to record extensively can generate speech that reflects their unique vocal characteristics, using existing voice samples.

2. Multilingual Speech Based on Japanese Voice Data

Using cross-lingual voice synthesis, the service enables speech in English, Chinese, French, Korean, and Spanish, all based on a user's Japanese voice data—enabling cross-language communication in one's own voice.

3. User-Centered Design Validated with ALS Patients

Over a 2.5-month period, the service was tested with 10 ALS patients, leading to iterative improvements. Enhancements include user settings, preset voice parameters, multilingual support, a stop button during playback, automatic text clearing after input, and smartphone interface optimization.

4. Accessibility Designed for Eye-Gaze Input

Buttons are large and well-spaced for eye-gaze navigation. Optional dark mode and high-contrast color schemes reduce eye strain and improve accessibility.

5. WITH ALS as a Supportive User Interface

WITH ALS provides end-to-end support—from onboarding consultation to ongoing usage—prioritizing empathetic, user-centered accompaniment beyond mere technical delivery.

Feedback from User Trials

Participants in the trial shared the following feedback:

"I'm glad it's easy to operate with eye-gaze input." "Since I can no longer speak, being able to recreate my voice from past recordings is truly invaluable." "I've always wanted to communicate in English—multilingual support is extremely convenient." "It's great that I can use it on my smartphone."

Demonstration at MOVE FES.2026

WITH ALS will host its 10th-anniversary event, 'MOVE FES.2026,' on June 20, 2026, at EX THEATER ROPPONGI. During the event, CEO Masaki Mutō will use 'Project XrossVoice' to deliver messages to the audience in his own synthesized voice.

Two major demonstrations are planned: First, during EYE VDJ MASA's live performance, he will energize the audience using Japanese and English vocal cues. Second, during the ALS TALK SHOW segment, he will use the service for real-time dialogue. This live showcase will allow the audience to experience how technology enables individuals to continue expressing themselves in their own voice, even after losing natural speech.

Service Overview

Service Name: Project XrossVoice

Launch Date: June 16, 2026

Target Users: Individuals who have lost or may lose their voice due to illness or other reasons

Service Type: Cloud-based voice synthesis communication service

Provider: WITH ALS, General Incorporated Association

Usage Model: Subscription-based

Supported Languages: Japanese, English, Chinese, French, Korean, Spanish

Pricing (tax included): 1-month plan: ¥4,500 / 3-month plan: ¥9,000 / 6-month plan: ¥15,000 / 12-month plan: ¥24,000

Registration Page: https://withals.com/category/?id=22