Response body

The body of the response is a JSON array.

Attribute

definition

type

Service type. Available services include:

*transcribe** - Speech to text
*translate** - Text to text translation
*audioalign** - Text to audio alignment
*languageid** - Language identification
*namedentity** - Named Entity Extraction
*sentiment** - Text sentiment detection
*texttospeech** - Text to speech

models

An array of model description objects.
name - The model name. E.g. “en-ru”.
description - The model description. E.g. “English to Russian”.
features - An array of strings identifying features available with this model

Model features

Models and services provide different features that can augment the data processing. The table below displays which features may be available for each of the services

Service	Feature	Description
Transcribe	punctuation	The transcript can include punctuation and true casing.
	advanced_srt_generation	The transcription can be saved in SRT format utilizing a specialized segmentation model which provides a more human-like subtitling.
	emotion	The transcription can include emotion information inferred from the audio features.
	sentiment	The transcription can include sentiment information.
	custom_vocabulary	User-defined custom vocabulary is supported.
	numeric_output	Numbers within the transcript can be converted to digit form rather than word form.
	domain	Identifies the domain that applies to this model.
Translate	advanced_srt_generation	When translating an SRT document, a specialized segmentation model can be utilized which provides a more human-like subtitling.
	meta_controls	The model is meta-aware and allows refinement of the translation output. See Query available translation models for more details.