Korean Dataset

한국어 데이터셋

High-Quality Korean Call-Center, Scripted Monologue, and Media (Podcast) Dataset for AI & Speech Models

Overview

Title (Language)

Korean Language Dataset

Dataset Types

Call Center, Media Data, Scripted Monologue

Country

Korea

Description

This dataset includes unscripted synthetic agent–customer telephonic conversations (5–15 minutes), licensable public domain audio/video files such as interviews and podcasts with 1 to 5 participants (15–60 minutes), and scripted monologues featuring a single speaker delivering predefined content for speech and language model training.

Use Case

ASR, Virtual Assistant, Chatbot, Conversational AI, Speech Analytics, TTS, Language Modelling

Data Set Details

Dataset Type Sampling Rate Speakers Channel Total Hours Total Number of Speakers
Call Center 8 kHz 2 Speakers Dual 102:27:55 1,244
Media Data 16 kHz Multiple Speakers Mono 203:46:54 397
Scripted Monologue 24 kHz Single Speaker Mono 500:00:00 On Request
Scripted Monologue 48 kHz Single Speaker Mono 1,955:00:00 On Request

Featured Clients

Empowering teams to build world-leading AI products.

Amazon
Google
Microsoft
Shaip contact us

Can’t find what you are looking for?

New off-the-shelf datasets are being collected across all data types

Contact us now to let go of your audio/speech training data collection worries

  • This field is for validation purposes and should be left unchanged.
  • By registering, I agree with Shaip Privacy Policy and Terms of Service and provide my consent to receive B2B marketing communication from Shaip.