This can be used to check if a word already exists in the base model lexicon or return a list of words that sound similar. Using the similar word list, the user can validate the system's phonetic generation.

The model name for which suggested words will be retrieved from

model

The words to retrieve similar sounding words for

Build a new user lexicon using the supplied word list.

The model id for which suggested words will be retrieved from

An optional descriptive name for this lexicon

name

The words and optional pronunciation guides (or sounds-like words)

Check on the status of the suggest or build request using the request ID received on submission of that request.

The request_id received when submitting the user lexicon request

request_id

Query the status of a Custom Vocabulary request and retrieve results

Return a list of user lexicons already built.

Use this method to remove a user lexicon.

The lexicon_id received when a user lexicon was successfully built.

lexicon_id

A simple HTTP service for submitting audio or text to determine the language within. For audio, the service returns a list of languages for each speech segment along with a confidence score. For text, the service returns a list of languages indentified with a confidence score for each. Data to be analyzed is provided within the body of the HTTP POST. It is important the that “Content-Type” is set correctly so that the correct engine is invoked for analysis. For audio, most popular audio and video file formats like the following are supported. mp3 - audio/mpeg, m4a - audio/m4a , ogg - audio/ogg , flac - audio/x-flac , wav - audio/x-wav or audio/wav. For text, only plain utf-8 encoded test is supported. txt - text/plain

The body of the POST contains the data to analyze.

Check on the status of request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a CSV text document in the format described below.
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

The request_id received when the request was submitted.

Query language identification request status and retrieve results

Use this method to remove request results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail.

Audio and text files are submitted via a HTTP multipart form POST. Each file is included as a multipart file attachment within the POST. The service supports several different formats for the audio file. The text must be a plain utf-8 encoded text file.

The name of the model to use. E.g. 'en.bcn'. You can use either the services endpoint or the audioalign/models endpoint or retrieve the available models.

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.

priority

The body of the request must contain text and audio file fields.

Returns a list of models available for audio alignment.

Check on the status of request which has been successfully submitted, using the request ID received on submission of that request.

If results are available, the body of the response will contain a document containing the output requested with the output query parameter (default is SRT).

srt - Primary clean SRT.
xml - A speech-to-text compatible XML.
xsrt - An SRT document which includes differences between untimed text reference and ASR transcript.
csv - A semicolon separated list of timed text segments with “INSERTION”, “SUBSTITUTION” and “DELETION” tags identifying difference between the untimed text and the ASR transcript.
html - A html formatted document with difference between the untimed text and the ASR transcript color coded.

If results are not available, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.


The type of output file to retrieve. eg. 'srt', 'xsrt', 'csv', 'html'

output

Query audio alignment request status and retrieve results

Use this method to remove request results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail.


The 'translation model' to use. You can use either the services endpoint or the translate/models endpoint or retrieve the available models.

Provide a list of meta controls to refine the translation output

meta_controls

The body of the POST contains the utf-8 text to be translated

Retrieve the available translation models.

Check on the status of request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a document in the same format as the document sent with the request.
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

The request_id received when the request was submitted

Query translation request status and retrieve results

Use this method to remove translation results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail.

Submit text to perform NLP Information Extraction and receive a tagged result with marked-up language constructs

The two-letter language identifier of the input text. E.g. 'en' for English.

language

Check on the status of the request which has been successfully submitted, using the request_id received on submission of that request

The request_id received when submitting the request

Query NLP Information Extraction request status and retrieve results

Use this method to remove request results.

Remove NLP Information Extraction results

Submit utf-8 text to perform Named Entity Identification

Check on the status of the request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a JSON object with entities labelled and extracted.
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

Query Named Entity Identification request status and retrieve results

Remove Named Entity Identification results

Submit a JSON array containing text for analysis.

The model id to use to synthesize speech. Use the services endpoint to determine the available languages

This method returns a list of available models/languages for sentiment analysis.

Check on the status of the request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a JSON object matching the input with sentiment score added to each segment.
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

Query Sentiment Analysis request status and retrieve results

Use this endpoint to remove request results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail.

This is a unified endpoint for all API services and will return a complete list of all services and models available for each

{
    "type"

"text"

"required"

"models"

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


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

{
    "punc"

punctuation will be added to the results (if available)."
    }

"diar"

speaker diarization annotations will be added to the results."
    }

"custom_vocabulary_id"

"channel_layout"

"sentiment"

"emotion"

Neutral

Angry

etc.)"
    }

"redact"

"priority"

"numeric_output"

The body of the POST contains the binary audio data.

This API allows the user to provide one or more urls to remote files to be downloaded and processed by the service.

Array of items with urls and configuration for each url

Transcribe an audio/video file from a remote url

Retrieve the available transcription models.

Check on the status of a request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a document in the format requested (default is XML).
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

Specify the result format to retrieve and optional formatting controls.

format

Query transcribe request status and retrieve results

Use this method to remove results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail

After retrieving a speech-to-text result in XML format, you can use this method to reformat the XML into one of several supported formats

Specify the ISO 639-1 two-letter language code for the input XML. E.g. “en” for English.

language_code

Use this method to submit text to be synthesized.

The name of the model to use to synthesize speech. E.g., 'en-US.default-en_male_35'. You can use the Query available text to speech models method to determine the available models

Use this method to submit text to be synthesized. The synthesize output will use features from the provided audio file to mimic the speech within the audio.
Audio and text files are submitted via a HTTP multipart form POST. Each file is included as a multipart file attachment within the POST. The service supports several different formats for the audio file. The text must be a plain utf-8 encoded text file.

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

The multipart request body containing both text and audio file fields.

Perform text to adapted synthesize speech

This method returns a list of available models that can be used for text to speech

Check on the status of the request which has been successfully submitted, using the request_id received on submission of that request.

If results are available, the body of the response will contain a WAV file with the synthesized speech.
Otherwise, HTTP status code 204 is returned with an empty response. Use the x-status header in the response to determine the request status.

Query text to speech request status and retrieve results

Use this method to remove text to speech request results.

IMPORTANT: Once results are removed, subsequent requests to retrieve the results will fail.

Apptek

Apptek API

Pre-configured, Predictable, Single Interface, Ready-to-use, Opinionated & Managed API Explorer & Server. Bring your API credentials, configure and access your 3rd party APIs. Using LowCodeAPI securely manage all integrations in one place.