KsponSpeech

by AI Hub Korea / ETRI

1000h Korean spontaneous-speech corpus — the open KR ASR baseline.

TL;DR

1000h Korean spontaneous-speech corpus — the open KR ASR baseline.

Best for korean ASR training and evaluation. Pricing: research-only.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
research-only
Platforms
AI Hub

What it is

KsponSpeech is ~1000 hours of spontaneous Korean dialogue across diverse topics. The standard open Korean ASR corpus, distributed via AI Hub Korea.

Best for: Korean ASR training and evaluation.
Watch out for: AI Hub license · KR-account-required · non-commercial restrictions apply · 2000 speakers. Cite: Bang et al., Interspeech 2020.

Install / use

https://aihub.or.kr/  # AI Hub account required

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported1
HIPAA eligibleNo

KsponSpeech vs Whipscribe

FeatureKsponSpeechWhipscribe
CategoryOpen sourceTranscription APIs
Pricingresearch-onlyfree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsAI HubWeb, API, MCP

Alternatives to KsponSpeech

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.