6) Dataset Info

Public Speech Dataset

Link: AI Hub — Public Speaking Practice & Assessment Data

Overview

This dataset was constructed to support public speaking practice and assessment. It contains presentation videos and speech audio, presentation text materials, and evaluation text data. The goal is to enable research and development in:

• Public speaking recognition and classification
• Speaking level evaluation

---

Speaker Distribution

Code	Group	Count	Ratio
A00	Middle school (Grade 9)	100	12.5%
A01	High school students	200	25%
A02	20s	200	25%
A03	30s	100	12.5%
A04	40s	100	12.5%
A05	50+	100	12.5%

---

Dataset Contents

• Length: 3~4 min per presentation
• Speakers: Metadata includes age group, gender, occupation, and audience type.
• Presentations: Topic, type, location, script text, and difficulty level.
• Utterances: Speech segments with start/end time, syllable count, word count, sentence count, and STT-based transcription.
• Metadata: File ID, filename, evaluation date, and data format.

---

Speech Presentation Dataset Schema

Category	Field	Description	Type
Speaker Info	speaker	Speaker ID	String
	age_flag	Age group	String
	gender	Gender	String
	job	Occupation	String
	aud_flag	Audience group	String
Presentation Info	presentation	Presentation ID	String
	presen_topic	Presentation topic	String
	presen_type	Presentation type	String
	presen_location	Presentation location	String
	presen_script	Original presentation script	String
	presen_difficulty	Presentation difficulty	String
Utterance Script	script	Utterance ID	String
	start_time	Utterance start time	String
	end_time	Utterance end time	String
	script_stt_txt	Utterance content (ASR/STT result)	String
	script_tag_txt	Utterance content (tag-mapped)	String
	syllable_cnt	Number of syllables	Number
	word_cnt	Number of words (tokens)	Number
	audible_word_cnt	Number of words clearly perceived by listener	Number
	sent_cnt	Number of sentences	Number
Evaluation	evaluations	Evaluation entry ID	String
	evaluation.eval_id	Evaluator ID	String
	eval_flag	Evaluator type	String
	eval_grade	Overall evaluation grade	String
Repetition	repeat_cnt	Count of repetitions/self-repairs	Number
	repeat_scr	Repetition/self-repair score	Number
Filler Words	filler_words_cnt	Count of fillers (um, uh, etc.)	Number
	filler_words_scr	Filler word score	Number
Pause	pause_cnt	Count of pauses	Number
	pause_scr	Pause score	Number
Pronunciation	wrong_cnt	Count of pronunciation errors	Number
	wrong_scr	Pronunciation score	Number
Voice Quality	voc_quality	Voice quality label	String
	voc_quality_scr	Voice quality score	Number
Voice Speed	voc_speed	Speech rate (words/sec)	Float
	voc_speed_sec_scr	Speech rate score	Number
Tagging	taglist	Tag list	String
	tag_id	Tag ID	String
	tag_keyword	Tag keyword	String
	tag_type	Tag type	Integer
Averages	repeat_scr	Average repetition/self-repair score	Float
	filler_words_scr	Average filler word score	Float
	pause_scr	Average pause score	Float
	wrong_scr	Average pronunciation score	Float
	voc_quality_scr	Average voice quality score	Float
	voc_speed_sec_scr	Average speech rate score	Float
	eval_grade	Average overall evaluation grade	String
Meta	info.filename	File name	String
	id	File ID	String
	date	Evaluation date	String
	formats	Data format	String

Participated Organizations

Organization	Responsibility
HealthCloud Co., Ltd.	Non-verbal data refinement & processing
GNUSoft Co., Ltd.	Linguistic AI modeling
ANeut Co., Ltd.	Non-verbal processing, AI modeling, authoring tools

---

Curriculum-Level Subject Dataset

• Source: AI Hub - Curriculum-Level Subject Dataset

• Overview This dataset is designed to support research in curriculum-aligned natural language understanding and multimodal learning. It was constructed through the systematic collection of textual and visual data from official educational materials, such as textbooks and reference guides, across multiple educational stages. These resources were then rigorously annotated and aligned with the achievement standards defined in the 2022 Revised National Curriculum of Korea, across nine core subject domains. The resulting dataset facilitates a range of educational AI tasks, including curriculum-based content inference, standard-level classification, and subject-specific knowledge modeling.

• Subjects: Science, Korean, Mathematics, English, Social Studies, Sociology, Ethics, Technology–Home Economics, Information (9 subjects in total)

• Preprocess: The dataset is partitioned into training and validation sets, each containing 80 textual samples per achievement standard to ensure balanced representation across labels.

• Distribution:
  After preprocessing, we collected a total of 1,071 achievement standards, each paired with 80 sample texts—resulting in 85,680 samples overall.

  Subject	Number of Standards	Total Samples
  Science	190	15,200
  Korean	209	16,720
  Technology and Home Economics	86	6,880
  Ethics	21	1,680
  Social Studies	173	13,840
  Society and Culture	13	1,040
  Math	241	19,280
  English	84	6,720
  Informatics	54	4,320
  Total	1071	85,680

• Contributors: Media Group Sarangwasup Co., Ltd.