Korean speech dataset, roughly 500 hours ready to license
Our OTS Korean speech dataset is roughly 500 hours of Korean speech, prompts and conversations, already recorded by Korean speakers in our 30,000+ network, with consent from every speaker for commercial AI training. License it now and train your Korean speech recognition model without waiting on a new collection.
- ~500 h of Korean in stock
- ~35,000 h across the OTS pool
- 100% of speakers consented
- ISO 27001 certified
- SOC 2 report under NDA
- GDPR DPAs and HIPAA BAAs on request
What's in the Korean speech dataset?
The IndecalAI OTS Korean speech dataset is roughly 500 hours of Korean speech, prompts and conversations, recorded by Korean speakers in our network who all consented to commercial AI training. It's already recorded, so you can license it now instead of waiting on a new collection.
South Korea has one of the highest rates of voice assistant and smart speaker use in the world, and Korean users expect polite, natural speech. Its sound changes and spacing rules are why Korean speech data needs Korean transcripts.
Every recording comes with a transcript and metadata about the speaker, the device and the setting, and you can take it as WAV, OGG or another common format. Every order starts with sample clips and a spec sheet, so you can match the data against your model before you agree a licence.
Need more than what's in stock? Korean speakers from our 30,000+ network record new speech to your brief, and our Korean data page covers what we collect to order.
Korean speech dataset at a glance
- Hours
- Roughly 500 hours, already recorded
- Language
- Korean (한국어), ko
- Content
- Speech, prompts and conversations
- Consent
- Every speaker consented to commercial AI training
- Speakers
- Korean speakers from our 30,000+ network
- Audio and transcripts
- Transcripts and metadata with every recording, in WAV, OGG and more
- Licence
- Terms agreed with each customer, in writing
- Price
- Quoted per order
- Samples
- Korean sample clips sent in reply to your request
- Security
- ISO 27001 certified. SOC 2 report available under NDA. GDPR DPAs and HIPAA BAAs on request.
What comes with every Korean recording
Recorded on the devices and in the places your users actually speak, and delivered in the format your pipeline expects.
| Detail | In the Korean dataset |
|---|---|
| Speech types | A mix of scripted prompts and natural conversations |
| Recording devices | Mobile phones, laptops and studio microphones |
| Environments | Quiet rooms and noisy, everyday settings |
| Transcripts | Included with every recording |
| Speaker metadata | Speaker details, such as gender and age, for each recording |
| Recording metadata | The device and the setting for each recording |
| Audio formats | WAV, OGG and other common formats |
| Sample rates | Listed for each recording on the spec sheet |
What roughly 500 hours of Korean speech looks like
Put end to end, 500 hours of Korean speech would play for about three weeks, day and night. Nobody listens to all of that before buying, so every order starts with sample clips and a spec sheet instead.
- ~500 hours of Korean speech
- ~30,000 minutes
- ~1.8M seconds of audio
- ~21 days, played back to back
Hours here are recorded audio. When you compare quotes, check whether each one counts audio hours or only the parts where someone is talking.
Korean at a glance, for speech teams
The facts that shape a Korean speech project, from who speaks it to how it's written. Our Korean data page covers what we collect to order.
Speakers
~82 million
people speak Korean, nearly all of them natively (Ethnologue, 2026).
Where it's spoken
South and North Korea
Spoken in both Koreas, with large communities in China, the US and Japan.
Script
한글 Hangul
An alphabet grouped into syllable blocks, with spaces between word groups.
Family
Koreanic
A Koreanic language with no close relatives outside Korea.
Varieties
Seoul and regional
Standard Korean is based on Seoul speech, and Gyeongsang, Jeolla and Jeju speech differ in sounds and words.
Codes
ko · kor
ISO 639-1 and ISO 639-3 codes. Locales speech APIs often use: ko-KR.
Korean speech dataset, voice dataset or speech corpus?
The same Korean data goes by several names, depending on who's asking. Here's what each name usually means.
Korean speech dataset
The general term
Recorded Korean speech with details about each recording, for training and testing speech models. The name this page uses.
Korean voice dataset
Usually the same thing
Some teams mean voice for text-to-speech or speaker recognition, which needs cleaner audio. Part of our data was recorded in a studio, so ask for those recordings.
Korean audio dataset
A wider term
It can include music or background sound. Ours is speech audio: Korean speech, prompts and conversations.
Korean speech corpus
The research name
A structured collection of recordings and their details. Commercial datasets are corpora too, licensed instead of published.
Korean ASR or speech-to-text dataset
Speech plus what was said
For training a model that turns Korean audio into text. Every recording in our Korean dataset comes with a transcript.
Korean voice data for assistants
Polite and casual speech
Assistants hear both polite and casual Korean. Ask which speech levels are in the hours you license.
What makes Korean hard for speech recognition
These are the things a Korean speech model trips on, and what to listen for in any Korean sample, ours included.
English loanwords in Korean
Korean uses many English loanwords, said the Korean way and written in Hangul, like 앱 for app. Transcripts have to spell them consistently, so check how a sample handles them.
Spaces and sound changes
Korean spacing rules are complex and often broken in typed text, and letters change sound next to each other. Check that the transcripts follow one spacing and spelling standard.
Accents across Korea
Seoul, Gyeongsang, Jeolla and Jeju speech differ in pitch, sounds and words. Check that the speaker mix covers the regions your users come from.
Speech levels
Korean verb endings change with politeness, so a request to a customer and to a friend end differently. A voice product needs speech in the levels its users speak.
What teams build with a Korean speech dataset
Common uses for Korean speech that's already recorded. They show the kind of work, and they aren't client names.
Korean speech recognition
Train or fine-tune a Korean speech-to-text model, or adapt a multilingual model so it handles Korean well.
Voice bots and IVR
Teach a voice bot how Korean callers actually ask for things, in the words and accents they use.
Customer service calls
Korean speech for models that route, transcribe or summarize customer calls about accounts, orders and claims.
Voice in the car and at home
A Korean base for in-car assistants and smart devices, before recording commands in the real setting.
Support and speech analytics
Korean conversations for models that find intent, sentiment or topics in support calls and chats.
Testing a Korean model
Held-back Korean recordings your model never trained on, to measure word and character error rate before launch.
How many hours of Korean speech do you need?
It depends on the model you start from and how close your users sound to the training data. These are rough starting points that come up often in speech work. Your own tests should decide.
| Goal | Rough starting point | With our Korean data |
|---|---|---|
| Test a model you already have | A few hours your model never trained on | A test set cut from stock and kept out of training |
| Fine-tune a multilingual model, such as Whisper or wav2vec 2.0 | Tens to a few hundred hours | Roughly 500 hours in stock covers it |
| Adapt to one domain, like banking calls | A smaller set that sounds like your users | Stock as a base, plus a custom top-up in your domain |
| Train a Korean model from scratch | Thousands of hours | Stock as a base, plus a larger custom collection |
Free Korean speech datasets or a licensed one?
Public corpora such as KsponSpeech, Zeroth-Korean, Mozilla Common Voice and FLEURS have done a lot for Korean speech research. They cover what their creators chose to record, like read sentences, crowd recordings or broadcast speech, and each comes with its own licence and terms of use. Here's how they compare with a licensed dataset.
| Topic | Public Korean corpora | Our Korean speech dataset |
|---|---|---|
| Licence | Set by each corpus, and some limit redistribution or changes | Agreed with you in writing, for what you're building |
| Consent | Covered by each project's own terms | Every speaker consented to commercial AI training |
| What it sounds like | Whatever the project recorded | Speech, prompts and conversations, with sample clips first |
| Who else trains on it | Anyone who downloads it | Set in the licence we agree with you |
| Filling gaps | Up to you | Recorded to your brief by the same network |
| Support | Community forums | A named contact from sample to delivery |
How to check a Korean speech sample
Thirty minutes with the sample clips and the spec sheet tells you most of what you need. Use the same list on any vendor.
| Check | What to look for | Where to find it |
|---|---|---|
| Accents | Speakers from the regions your users come from | The speaker mix on the spec sheet, and the clips |
| Everyday speech | The way your users really talk, including any mixing with other languages | Listen to the clips |
| Script | Transcripts in the script and spelling your model should output | The transcript format on the spec sheet |
| Audio | A sample rate and recording setup close to your product | The sample rate on the spec sheet |
| Numbers | Amounts, dates and phone numbers written one consistent way | Transcripts of the clips |
| Rights | Consent that covers commercial AI training, and a licence that fits | Ask for the licence terms before you buy |
From Korean sample clips to licensed data
Your named contact runs it from the first question to delivery.
-
01
Ask
Tell us how many Korean hours you need and what the model is for.
You get A reply within one business day
-
02
Sample and spec sheet
We send Korean sample clips and the spec sheet in reply.
You get Korean clips and the spec sheet
-
03
Licence and quote
We agree the licence terms and send a quote for the order.
You get Written licence terms and a quote
-
04
Delivery
The licensed Korean speech is delivered to you.
You get Licensed Korean speech with transcripts
-
05
Fill the gaps
Accents, devices or domains that are missing get recorded to your brief.
You get New Korean speech, recorded to your brief
Get Korean sample clips Start with the sample clips, then decide.
Korean speech data terms
The words that come up when you buy or build Korean speech data. The full glossary covers every data type.
- Hangul
- The Korean alphabet, with letters grouped into syllable blocks such as 한.
- Speech level
- The politeness level shown by a Korean verb ending, from casual to formal.
- Liaison
- A final consonant carrying over to the next syllable in speech, so Korean words sound different from how they're spelled.
- Speaker metadata
- Details recorded about each speaker, such as gender and age, that let you check the speaker mix against your users.
- OTS dataset
- Short for off-the-shelf dataset: data already collected and ready to license before a buyer asks for it, instead of recorded to one buyer's brief.
- Word error rate (WER)
- The share of words a speech model gets wrong against a reference transcript. Lower is better.
- Character error rate (CER)
- The same idea counted in characters, often used alongside WER, since small spelling differences can make WER look worse than the model is.
- Utterance
- One stretch of speech by one speaker, such as a single prompt read aloud or one turn in a conversation.
More Korean and speech data
Every language in stock is on the off-the-shelf speech datasets page. Related speech datasets: Japanese, Mandarin Chinese, English and Vietnamese.
OTS speech datasets
Around 35,000 hours across languages, ready to license.
Korean data to your brief
Korean speech, text and translation, collected to order.
Transcription
Korean audio turned into timed text, checked by people.
Languages
100 languages, collected to order.
Data and model evaluation
Held-back Korean speech scored against your targets.
Questions about the Korean speech dataset
Straight answers about licensing Korean speech. The off-the-shelf FAQ covers questions that apply to every language.
Do you have a Korean speech dataset ready to license?
Yes. Roughly 500 hours of Korean speech, prompts and conversations are already recorded and ready to license, as part of our off-the-shelf speech pool.
What's in the Korean speech dataset?
A mix of scripted prompts and natural conversations in Korean, recorded on mobile phones, laptops and studio microphones, in quiet and noisy settings. Every recording comes with a transcript and metadata about the speaker, the device and the setting, in WAV, OGG or another common format.
How are the Korean transcripts spaced?
Ask for the transcript style on the spec sheet, including spacing, numbers and loanwords. You can check it against your pipeline before you license.
Did every speaker consent to AI training?
Yes. Every speaker gave consent for their recordings to be used for commercial AI training.
How is the Korean dataset licensed?
Licence terms are agreed with each customer, in writing, before delivery. That includes how you can use the data and whether anyone else can license the same recordings.
How much does the Korean speech dataset cost?
We quote each order on its own, based on the hours you need and the licence terms. Send your request and your named contact comes back with a quote.
Can we hear Korean samples first?
Yes. Ask and we send Korean sample clips with the spec sheet, so you can check the audio before you commit.
What if we need more than 500 hours, or a specific accent?
We collect it. Korean speakers from our network record new speech to your brief, such as a region, an age group, a device or a domain, and it can sit alongside the hours you license from stock.
Is this the same as a Korean voice dataset or speech corpus?
For most uses, yes. Speech dataset, voice dataset, audio dataset and speech corpus are names for the same thing: recorded Korean speech with details about each recording. If you need studio-quality voice for text-to-speech, ask for the recordings made in a studio.
Can we use it as a Korean speech-to-text (ASR) dataset?
Yes. Every recording comes with a transcript, so it's ready for training and testing Korean speech-to-text models. The spec sheet shows the transcript style, so you can check it matches your pipeline before you license it.
Why not use a free Korean speech dataset?
Public Korean corpora are a good start for research. Each one has its own licence and terms of use, and they cover what their creators chose to record, such as read sentences or broadcast speech. Read the licence before you train a commercial model on one, and check that it sounds like your users.
What does OTS mean?
OTS stands for off-the-shelf: data that's already recorded and ready to license, instead of collected after you order. Our OTS Korean speech dataset is roughly 500 hours.
How many hours of Korean speech do we need?
It depends on your starting model. A test set can be a few hours, fine-tuning a multilingual model often starts at tens to a few hundred hours, and training from scratch takes thousands. Your own tests should decide.
Can we fine-tune Whisper or wav2vec 2.0 on it?
Yes. Teams commonly fine-tune open speech models like these on licensed Korean speech, and every recording comes with a transcript in WAV, OGG or another common format. The spec sheet shows the transcript style, so you can plan any conversion before you train.
Do you work with US teams?
Yes. We take calls in US Eastern mornings, which is evening in India. We're ISO 27001 certified, our SOC 2 report is available under NDA, and we sign GDPR data processing agreements and HIPAA business associate agreements on request.
Get Korean sample clips
Say roughly how many Korean hours you need and what the model is for. We'll send sample clips and the spec sheet.
- Your named contact replies within one business day.
- We send Korean sample clips and the spec sheet.
- We agree the licence terms and send a quote.
This opens your email app with the request filled in. Nothing is sent until you press send. You can also use the contact page.