Next-Generation English Conversation Learning English Deep Learning" to be Launched on April 1 ～Inspired by the brain's speech prediction mechanism

Deep Learning, LLC (Location: Fujimino-shi, Saitama, CEO: Keizo Sakurai), an English conversation learning company, will start offering a new service "Deep Learning of English" on April 1, 2026 (Wednesday) for effective learning for English conversation learners.

English fluency for 700 yen/month

Many learners have this problem: they can understand English, but the words don't come out at a moment's notice.

To address this issue that many learners face, a new English training program has been developed that is inspired by the way the brain works when speaking.

It is said that when both children and adults speak, they predict their own voice in their brains before they speak, and subconsciously correct any discrepancies between their voice and the actual sound. This mechanism of speech prediction and feedback is known in motor control research and has been discussed in the field of neuroscience.

This program was designed with the idea of "prediction and self-correction" in native language acquisition in mind. Through repeated imitation of native speech and self-feedback, the program aims to build a memory that can speak without translation by understanding English in the language of thought.

The program has been redesigned for adult learners, referring to the automated process found in native language acquisition.

At the affordable price of 700 yen/month, the system helps students retain natural, native-like English without relying on conventional "translation.

URL for "Deep Learning of English": https://ディープラーニング.net/

Background of development: Why can't Japanese people speak English?

Many English learners feel that they can understand but cannot speak at a moment's notice. This is thought to be related to the fact that they compose sentences based on grammar and vocabulary, and that semantic processing and expression generation are not sufficiently automated at the time of speech.

This service promotes the integration of effective speech learning and speech motor skills through a series of exercises that combine speech and auditory feedback, just like in the native language.

The adult brain also has plasticity and the ability to modify, but in the case of a second language, it is optimized for the native language and its interference inhibits automation. This program aims to reduce translation dependence and strengthen the linkage between semantic processing and speech production in English, as in native language acquisition, while overcoming the interference of the native language.

■ Distinctive mechanisms of this service

1. speech learning utilizing speech prediction and sensory feedback

Neither native nor second language speakers use grammar or word knowledge when speaking a language.

The brain functions with motor commands as well as collateral discharge, which predicts the resulting sensory information. This is a widely studied neural mechanism that automatically corrects itself with feedback after imitation is learned. Through this mechanism, language production (speaking) and perception (listening) are trained in tandem, efficiently promoting conversational automaticity.

Adults who have passed the critical period have difficulty learning a language not because that function does not work, but because that function is optimized for the native language and is subject to its interference.

This service aims for effective learning like native language through imitation of native speech by understanding speech movements in thought language, auditory feedback of bone-conducted speech, and consciously overcoming the interference of the native language by reducing the error between prediction and actuality.

The design concept of overcoming the causes of inability to speak even if understood is distinct from conventional input-oriented learning.

2. utilization of self-speech feedback

Self-speech is perceived through both air-conducted and bone-conducted sound. In speech learning research, it is known that bone-conducted speech, the process of constantly listening to and correcting one's own speech, is important.

In this service, the participants practice carefully checking their own speech after utterance with bone-conduction speech, taking advantage of the plasticity of the adult brain, and noticing differences from model speech.

It has been suggested that this service facilitates the coordination of speech movements, the reconstruction of phonological categories, and the integration of perception and production.

3. imitation learning and maintenance of motivation

Language acquisition studies have reported that repeated imitation and meaningful use contribute to retention. In addition, a sense of accomplishment and improvement are also considered important psychological factors for continued learning.

This service aims at natural language learning, emphasizing overcoming the interference of the native language through skill improvement by imitation utilizing the plasticity of the brain, mutual support in the community, and pricing that makes it easy to continue. 4.

4. learning design that includes not only "listening" but also "prediction

In recent years, cognitive neuroscience has widely supported the "predictive encoding model," in which the brain does not process sounds passively but actively processes input while generating predictions.

Speech and listening practice may help form internal models that predict speech sounds. This service uses a design that takes advantage of the plasticity of the brain to enhance speech prediction ability through speech.

The service predicts English in the language of thought, allowing for natural English conversation without the need to translate Japanese.

Participation of Researchers Related to Language Acquisition

Since this site is a new English learning method that is still in the process of being put into practical use, researchers interested in cognitive studies, second language acquisition, linguistics, and the reproduction of native language mechanisms are welcome to participate and verify the effectiveness of this method, and to practice or conduct research on it.

Future Prospects

Deep Learning for English" will transform the "continuous effort" of language learning into "an enjoyable experience where you can feel your progress" by combining technology and brain science, aiming to provide an environment for effective autonomous learning like AI, rather than interacting with AI.

■Service Overview

Service name: English Deep Learning

Date of launch: April 1, 2026

Using the materials on the above site, students will learn English conversation by imitating native speakers in the same way as they do in their native language.

Subscription: 770 yen per month (tax included), 7,700 yen per year (tax included)

Dialogues: 60 dialogues (approx. 3,000 English sentences)

Expressions: 60 books (3,000 English sentences)

English learning community through DISCORD

Support by YouTube videos

Main learning contents: Imitate native speakers by understanding English in the language of thought as if it were your native language.

　　　　　　　　 Correction by internal feedback of bone-conducted speech without devices

Company Profile

Company Name ： Limited Liability Company "Deep Learning

Representative: Keizo Sakurai, Representative Partner

Location ： 4-4-19-502 Kamifukuoka, Fujimino-shi, Saitama 356-0004, Japan

Establishment ： June 2016

Business : English language education

Capital ： 100,000 yen

URL ： https://ディープラーニング.net/

For customer inquiries regarding this matter, please contact

Deep Learning, LLC Customer Service Center

TEL: 049-265-1229