STT (Speech-To-Text)

Minds Lab’s STT API transforms speeches into texts

Getting Started

① Input: Sound file (.wav/.pcm/.mp3)

② Model: Choose one of below

  • Korean 8K (baseline, kor, 8000)
  • Korean 16K (baseline, kor, 16000)
  • English 8K (baseline, eng, 8000)
  • English 16K (baseline, eng, 16000)
  • Korean Address (address, kor, 8000)
  • English for Localized Korean (kor_accent, eng, 8000)
  • English for Teachers (mindsedu_t, eng, 8000)
  • English for Students (mindsedu_s, eng, 8000)
Start API Service

① Request

  • Method : POST
  • URL :

② Request parameters

Key Explanation type
ID Unique ID. Ask us for the ID string
key Unique key. Ask us for the key string
cmd Keyword for the API (runFileStt) string
lang Language of the model ( kor / eng ) string
sampling Sampling rate of the model (8000 / 16000) string
level Level of the model (baseline / address / kor_accent / mindsedu_t / mindsedu_s ) string
file Type: Sound file (wav / pcm / mp3) string

③ Request example

curl -X POST \ \
-H 'content-type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW' \
-F ID= (*Request for the ID) \
-F key= (*Request for the key) \
-F lang=eng \
-F sampling=8000 \
-F level=baseline \
-F 'file=@eng_8k_1.wav' \
-F cmd=runFileStt

④ Response parameters

Key Explanation type
status Status of the API service (Success / Fail) string
data Result: Body text string
extra_data[‘stt_duration’] Extra data: Duration of the sound file (second) string
extra_data[‘stt_length’] Extra data: Length of the result text (characters) string

⑤ Response example

"status": "Success",
"extra_data": {
"stt_length": 18,
"stt_data": "Let's play badminton.\n",
"stt_duration": "2.387"
"data": "Let's play badminton.\n"

Contact Us

Have questions on MINDs Lab or MINDs Lab’s API? Please ask us.
Also, please follow us on Facebook and Linkedin for recent updates and events.

Contact Us