Visual Bank Inc. (Minato-ku, Tokyo; CEO: Saneyuki Nagai), through its subsidiary amanaimages Inc., has launched a new dataset under its AI training data solution, Qlean Dataset: the Japanese Single-Speaker Kodan Speech Corpus with Transcripts. The dataset is designed for speech- and language-based AI development and research, including Automatic Speech Recognition (ASR), speech understanding, and speech-language modeling.

This dataset consists of audio recordings of Kodan, a traditional Japanese narrative storytelling performance, delivered by a single speaker. Each recording is paired with a Japanese transcript that faithfully reflects the spoken content. The corpus captures continuous natural speech that includes expressive intonation, pauses, and variations in speaking rate characteristic of Kodan performance. Unlike read-aloud or conversational datasets, this corpus contains narrative speech structures specific to Japanese storytelling.

Because Kodan narration incorporates scene descriptions, character differentiation, and dramatic tension as the story unfolds, the dataset provides a structured environment for examining the alignment between acoustic signals and textual expression. It enables evaluation under expressive, non-monotonic speech conditions that differ from standard scripted reading datasets. With recordings ranging from short to long form, the corpus also supports research on contextual retention and segmentation in continuous speech processing.

In response to data requirements in both research and commercial AI development, including foundation model development, Qlean Dataset provides this corpus with clarified usage rights and licensing conditions. Visual Bank will continue expanding structured Japanese-language datasets in the speech and language domain to support the foundation of AI development and research.

 Overview of the Japanese Single-Speaker Kodan Speech Corpus with Transcripts

Use Case Examples 

Research Applications 

• Evaluation of Natural Speech Recognition Accuracy in Japanese ASR Models
  The corpus enables evaluation of ASR models using continuous narrative speech that includes expressive intonation and pauses, allowing analysis of recognition accuracy and error patterns under conditions that differ from standard read speech.

• Research on the Relationship Between Acoustic and Linguistic Representation
  By combining audio signals with aligned transcripts, researchers can analyze how narrative structure and prosodic features in Japanese influence language understanding.

 Industry Applications 

• Validation of Long-Form Speech Processing in Voice-Enabled AI Systems
  For AI products involving voice search or audio archive analysis, the dataset can be used to validate speech segmentation, full transcription, and summarization performance using extended monologue recordings.

• Pre-training and Evaluation of Japanese Speech-Language Models
  As a dataset containing narrative speech structures specific to Japanese, it can serve as supplementary data for pre-training or evaluation phases of speech-language models.

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 a wide range of data types, including images, videos, audio, 3D assets, 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 continues to expand its specialized, industry-focused lineup known as the “AI Data Recipe.”
By reducing the operational burden of data collection and preparation, Qlean Dataset helps organizations establish AI development environments that are both legally compliant and risk-free.

▶ 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

• Existing datasets deliverable within one business day

• Custom data collection and recording services 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 enhance AI development capabilities under the mission “Unlocking Data Accessibility.”
The company operates THE PEN, an AI-assisted creative tool for manga artists and the Qlean Dataset service.
Its subsidiaries include Amana Images Inc., one of Japan’s largest photostock providers; Qlean Dataset, which leads research and development in AI data; and THE PEN Inc., an AI-assisted creative tool for manga artists.

CEO: Saneyuki Nagai
Address: 6F, C-Cube Minami Aoyama Building, 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