Skip to main content
POST
Transcribe speech

Authorizations

Authorization
string
header
required

Your geko API key (sk-tokay-…). Create and manage keys in the console.

Headers

x-filename
string

Filename hint; only the extension matters, so the server knows the container. Defaults to audio.wav.

Example:

"call.wav"

Query Parameters

url
string<uri>

Public URL of the audio. Use instead of sending a body.

Body

The body is of type file.

Response

The transcript, with timing and the credits charged.

text
string
required

The transcript. Lowercase and unpunctuated — the model's vocabulary contains neither, so nothing is being stripped.

Example:

"сәлеметсіз бе бүгін ауа райы өте жақсы"

model
string
required
Example:

"seta-kk-ru-v2"

audio_seconds
number
required

Duration of the submitted audio. This is the billed quantity.

Example:

2.84

credits_charged
integer
required

Credits billed: ceil(audio_seconds × 2.5).

Example:

8

processing_seconds
number
Example:

0.153

rtf
number | null

Real-time factor: processing_seconds / audio_seconds. Lower is faster.

Example:

0.0539

x_realtime
number | null

How many times faster than real time this ran.

Example:

18.5

chunks
integer

How many pieces the audio was split into at pause boundaries.

Example:

1

sample_rate_in
integer

Sample rate of the submitted audio, before resampling to 16 kHz.

Example:

24000