Visual Bank Inc. (Minato-ku, Tokyo; CEO: Saneyuki Nagai) has launched a new dataset within its AI training data solution “Qlean Dataset,” provided through its subsidiary Amana Images Inc. The newly released Japanese Single-Speaker History-Themed Speech Corpus Dataset contains monologue recordings on topics spanning Japanese history, world history, and cultural history.
This dataset supports training and evaluation for automatic speech recognition (ASR), natural language processing (NLP), and foundational generative AI models.

This corpus consists of long-form Japanese monologue recordings spoken by male and female speakers in their 20s to 50s. The recordings feature natural, unscripted speech, preserving discourse elements such as context-dependent explanations, topic shifts, and episodic storytelling. All audio was recorded in MP3 format at 44.1 kHz, totaling roughly 150 hours across clips ranging from 5 to 40 minutes. The dataset is suitable for advanced language-processing tasks requiring contextual understanding, summarization, and semantic inference based on speech input.

Because the dataset includes domain-specific narratives in the field of history, it is useful for long-form speech processing, context-aware ASR evaluation, and improving Japanese-language capabilities in generative AI models. Its natural conversational flow and topic development make it applicable for generalization testing, dialogue systems, educational AI tools, and other real-world model development.

Dataset Specifications

Use Case Examples Research

Use (Academia)

• Training and evaluation of long-form ASR models
  The context-dependent monologues containing historical terminology enable evaluation of recognition accuracy for long-duration audio and analysis of error patterns.

• Japanese NLP research (summarization, NER, discourse analysis)
  The dataset’s explanation structures and topic transitions support research in summarization, discourse analysis, and named entity recognition.

• Research on audio-to-text-to-reasoning pipelines for generative AI
  The continuous monologue format is suitable for multi-step AI models involving speech transcription, content understanding, and text generation.

Industry Use (Enterprise)

• Improving Japanese ASR accuracy
  The monologues contain specialized vocabulary and long-form structures conducive to strengthening ASR models in education, content, and knowledge domains.

• Enhancing knowledge-aware conversational AI and chatbots
  The dataset supports training dialogue systems that generate extended responses or structured explanations.

• Evaluation for speech-driven LLMs and multimodal AI
  Context-rich long-form recordings enable verification of the audio → text → reasoning workflow.

Other Use Cases (Education / Social Implementation)

• Developing AI systems for educational content generation
  By learning from historical explanatory speech, AI models can improve the quality of generated educational content such as explanations and summaries.

About Qlean Dataset

Qlean Dataset is a commercial-use-ready AI training data solution provided by Amana Images Inc., a subsidiary of Visual Bank Inc.
It supports diverse data types including images, videos, audio, 3D, and text—enabling both research and commercial AI development in a legally safe environment.

Through collaborations with data partners such as Chiba Lotte Marines Co., Ltd. and Toyo Keizai Inc., Qlean Dataset continuously expands its specialized, industry-relevant lineup known as the “AI Data Recipe.”

By reducing the operational burden of data collection and preparation, Qlean Dataset helps build legally compliant and risk-free AI development environments.

▶ Qlean Dataset: https://qleandataset.visual-bank.co.jp/en
▶ AI Data Recipe: https://qleandataset.visual-bank.co.jp/en/lineup

Key Features of Qlean Dataset

• Full consent obtained from all subjects; compliant with GDPR and CCPA

• Existing datasets deliverable within one business day

• Custom data collection and recording available

▶ Contact: https://qleandataset.visual-bank.co.jp/en/contact

About Visual Bank Inc.

Visual Bank Inc. is a Tokyo-based startup building next-generation data infrastructure to maximize AI development capabilities under the mission, “Unlock the potential of all data.”
The company operates THE PEN, an AI-assisted creative tool for manga artists, and wholly owns Amana Images Inc., which provides the Qlean Dataset service.

CEO: Saneyuki Nagai
Address: C-Cube Minami Aoyama Building 6F, 7-1-7 Minami-Aoyama, Minato-ku, Tokyo 107-0062
Corporate Site: https://visual-bank.co.jp/en/
Amana Images: https://qleandataset.visual-bank.co.jp/en/company-overview