Transcribe an audio/video file

Submit audio for transcription, The body of the POST contains the binary audio.

The Content-Type header should be set according to the media type. Most popular audio and video file formats are supported, including the following:

mp3 - audio/mpeg
m4a - audio/m4a
ogg - audio/ogg
flac - audio/x-flac
wav - audio/x-wav, audio/wav

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…

Channel Layout format

The format is [stream].channel[,[stream].channel] where stream identifies which media stream in the audio file to select and channel is the audio channel. Both values are zero based, so the first stream is ‘0’ and the first audio channel within the stream is ‘0’. If ‘stream’ is not defined, ‘0’ is assumed.

Examples:
channel_layout=0 selects the first audio channel within the first media stream.
channel_layout=0,1 selects the first two channels in the first media stream.

If this parameter is not specified, all audio channels from all media stream are processed.


User request completion callback

When submitting a request, an optional HTTP callback can be specified which will be invoked when the request completes or fails. The callback is an HTTP POST to the exact URL provided. Only one attempt is made to reach the URL provided. The contents of the HTTP request match the response from the “Querying a request status and retrieving results” above.

Headers

  • x-requestid - Contains the request_id received when the request was submitted
  • x-status - One of the following indicating the status of the request:
StatusDescription
"Completed"The request has finished processing and the user callback, if provided has been called.
"Failed"An error occurred processing the request.
  • x-error - If the request failed, this would contain an error message.
  • x-userdata - This will contain the user data supplied with the original request.

Body

If the status is "Completed" the body of the request contains the results from the translation. If the status is "Failed" the body of the request contains the same error message provided in the x-error header.

Details

Path Params
string
required

The ASR model to submit this request to. You can use either the services endpoint or the /transcribe/models endpoint or retrieve the available models.

Query Params
boolean
Defaults to false

If set to a non-zero value, punctuation will be added to the results (if available).

boolean
Defaults to false

If set to a non-zero value, speaker diarization annotations will be added to the results.

string

Set the user lexicon id to use during the recognition

string

This is used to define which audio channel(s) in the media file to transcribe. (see the Channel Layout section below)

boolean
Defaults to false

Include sentiment scores in the transcription. Sentiment values range from 0.0 (Negative) to 1.0 (Positive)

boolean
Defaults to false

Include emotion tagging (Happy, Neutral, Angry, etc.)

boolean
Defaults to false

Redact personal information from the transcript and the audio. Text redacted from the transcript is replaced with the identified class. E.g. "[_pers]" being a personal name.

int32
Defaults to 50

An integer priority level between 1 (low priority) and 100 (high priority) for this request. Additional charges may be applied to jobs with higher priority values. Default value is 50.

boolean
Defaults to false

Convert to numbers in word form to digit form. E.g. "two thousand three" converted to "2003"

string

Select a domain specific model. Use the 'services' or 'transcribe/models' API to identify available domains.

Body Params
string
required

The body of the POST contains the binary audio data.

Headers
string
required

API Key

string

(optional) A user HTTP callback URL which will be invoked to notify the user when the request completes. See “User request completion callback” for further description.

string

(optional) Any text data the caller wants to associate with this request. This can be used to link this request with some other external tracking information.

Responses

Language
Credentials
Header
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json