Handwriting OCR Data of Japanese and Korean | 100 People | Computer Vision Data
Image
100 People - Handwriting OCR Data of Japanese and Korean,.
- Language
- Japanese, Korean
- Volume
- 100 people
- Domain
- AI & Machine Learning +1
- Pricing
- USD 10,000
Language
The catalogue lists 125 Korean datasets. Most listings are text, audio and image data, typically used for LLM pre-training, document understanding & OCR and fine-tuning & instruction data. Where providers state a volume in hours, it ranges from 3 hours to 23.9k hours.
Regional variants listed: Korean (Korea (the Republic of)), Korean (ko).
Image
100 People - Handwriting OCR Data of Japanese and Korean,.
Audio · Text
2 People - Korean Average Tone Speech Synthesis Corpus.
Audio
Korean(Korea) Spontaneous Dialogue Smartphone speech dataset, collected from dialogues based on given topics, covering 20+ domains.
Audio
Korean(Korea) Spontaneous Dialogue Telephony speech dataset, collected from dialogues based on given topics, covering 20+ domains.
Audio
This 200 Hours Korean Full-Duplex Multi-Channel Speech Dataset features multi-stream audio recorded via smartphones, simulating natural conversations across a range of…
Audio
Korean(Korea) Real-world Casual Conversation and Monologue speech dataset, mirrors real-world interactions.
Video · Multimodal
This dataset contains 10,000 high-quality GUI interaction sequences (continuous frames) across mobile, tablet, and PC devices, covering scenarios such as desktop…
Audio
Mixed Speech with Korean and English Scripted Monologue Smartphone speech dataset, collected from monologue based on given prompts, covering oral category; human-machine…
Text · Medical
Annual health screens for 3,285 patients located in korea, about ~3 check ups per person on average, 9,187 reports / screens (csvs) in total in 15474 cvs files.
Text
This dataset features a curated collection of questions and answers from a q&a community platform, where users receive high-quality responses from verified experts…
Audio · Multimodal
Specifications: ‣ Volume: 100,000+ hours ‣ Format: 8kHz, mono channel (WAV audio + JSON transcript) ‣ Environment: Studio, podcast-quality recording ‣ Recording content:…
Image · Medical
Most healthcare datasets are take-it-or-leave-it: you get the cohort someone else already assembled.