Skip to main content

Speech to Text: get result

GET 

/api/technology/speech-to-text/:task_id

Retrieve the result of a Speech to Text task.

Request

Path Parameters

    task_id uuidrequired

    ID of the task to get result of

Header Parameters

    x-correlation-id X-Correlation-Id

    Correlation ID is a special type of request ID which is unique over a series of requests and responses, identifying a transaction in a distributed system. Correlation ID will be generated if not provided.

    x-request-id X-Request-Id

    In distributed system architecture (microservices architecture) it is a unique ID of request and response combination throughout all components of a distributed system. Request ID will be generated if not provided.

Responses

Result response of Speech to Text task.

The response body contains task info in the task field. Depending on the task processing outcome, either result or error field is populated.

Schema

    task

    object

    required

    task_id uuidrequired
    state TaskInfoState (string)required

    Possible values: [pending, running, rejected, failed, done]

    result

    object

    anyOf

    Result of Speech to Text task.

    one_best

    object

    required

    One-best result of Speech to Text task.

    segments

    object[]

    required

  • Array [

  • channel_number Channel Number (integer)required

    Channel number (0-base indexed).

    start_time Start Time (number)required

    Start time of the segment

    end_time End Time (number)required

    End time of the segment

    text Text (string)required

    Transcription of the whole segment

    words

    object[]

    required

    Detailed per-word segmentation of the segment

  • Array [

  • start_time Start Time (number)required

    Start time of the word (seconds)

    end_time End Time (number)required

    End time of the word (seconds)

    text Text (string)required

    Transcription of the word

  • ]

  • ]

  • additional_words

    object[]

    required

    Additional words extending the built-in vocabulary after having been processed by the technology. There are two possible results of processing. If you provided a spelling without an explicit pronunciation, the technology has generated a pronunciation for you. Alternatively, if you provided a new pronunciation for a word that was already part of the model's vocabulary, the result also contains built-in pronunciations along with the new one.

  • Array [

  • spelling Spelling (string)required

    The written form of the word described by graphemes.

    pronunciations

    object[]

    required

  • Array [

  • pronunciation Pronunciation (string)required

    The word's pronunciation described by phonemes.

    out_of_vocabulary Out Of Vocabulary (boolean)required

    If false, this pronunciation of the word is known to the language model. If true, it is a new pronunciation that has extended the model's abilities.

  • ]

  • ]

  • error

    object

    anyOf

    Result schema for a failed or rejected task.

    type TaskErrorType (string)required

    Possible values: [internal, task.validation-error]

    Machine-readable error type.

    message Message (string)required

    Human-readable summary of the error.

    detail

    object[]

    Optional higher level of detail. It is intended for better understanding of the error or advanced error handling.

  • Array [

  • location

    object[]

    required

    Location of the error.

  • Array [

  • anyOf

    integer

  • ]

  • message Message (string)required

    Human-readable summary of the error.

    type Type (string)required

    Machine-readable error type.

    context

    object

    Optional key-value object with additional context

    property name*

    object

    anyOf

    string

  • ]

Loading...