V2 API Reference

The Speechmatics Automatic Speech Recognition REST API is used to submit ASR jobs and receive the results. The supported job type is transcription of audio files.

Version: 2.7.0

Terms of service

https://www.speechmatics.com/terms-and-conditions/

Contact information

support@speechmatics.com

URI scheme

BasePath: /v2/jobs/

Schemes: HTTPS, HTTP

Paths

The base URL https://${APPLIANCE_HOST}/v2/jobs/ is used for REST Speech API requests. If you are using HTTP, the base URL is: http://${APPLIANCE_HOST}:8082/v2/jobs/.

/jobs

Requests without a job ID component are used to create a new job, or to return a list of all submitted jobs

POST

Summary: Create a new job.

Parameters

Name	Located in	Description	Required	Schema
config	formData	JSON containing a JobConfig model indicating the type and parameters for the recognition job.	Yes	string
data_file	formData	The data file to be processed. Alternatively the data file can be fetched from a url specified in JobConfig.	No	file

Responses

Code	Description	Schema
201	OK	CreateJobResponse
400	Bad request	ErrorResponse
401	Unauthorized	ErrorResponse
403	Forbidden	ErrorResponse
500	Internal Server Error	ErrorResponse

GET

Summary: List all jobs.

Responses

Code	Description	Schema
200	OK	RetrieveJobsResponse
401	Unauthorized	ErrorResponse
500	Internal Server Error	ErrorResponse

HTTP Method GET

Summary: Get job details, including progress and any error reports.

/jobs/{jobid}

Requests with a job ID component are used to view the status, transcript or audio data for a job, or remove a given job from the system.

Parameters

Name	Located in	Description	Required	Schema
jobid	path	ID of the job.	Yes	string

Responses

Code	Description	Schema
200	OK	RetrieveJobResponse
401	Unauthorized	ErrorResponse
404	Not found	ErrorResponse
500	Internal Server Error	ErrorResponse

HTTP Method DELETE

Summary: Delete a job and remove all associated resources.

Parameters

Name	Located in	Description	Required	Schema
jobid	path	ID of the job to delete.	Yes	string
force	query	When set, a running job will be force terminated. When unset (default), a running job will not be terminated and request will return HTTP 423 Locked.	No	boolean

Responses

Code	Description	Schema
200	The job that was deleted.	DeleteJobResponse
401	Unauthorized	ErrorResponse
404	Not found	ErrorResponse
423	Locked	ErrorResponse
500	Internal Server Error	ErrorResponse

HTTP Method GET

/jobs/{jobid}/transcript

Summary: Get the transcript for a transcription job.

Parameters

Name	Located in	Description	Required	Schema
jobid	path	ID of the job.	Yes	string
format	query	The transcripton format (by default the json-v2 format is returned). txt and srt are also supported/	No	string

Responses

Code	Description	Schema
200	OK	RetrieveTranscriptResponse
401	Unauthorized	ErrorResponse
404	Not found	ErrorResponse
410	Gone	ErrorResponse
500	Internal Server Error	ErrorResponse

HTTP Method GET

/jobs/{jobid}/log

Summary: Get the log file for a transcription job.

Parameters

Name	Located in	Description	Required	Schema
jobid	path	ID of the job.	Yes	string

Responses

Code	Description	Schema
200	OK	file
401	Unauthorized	ErrorResponse
404	Not Found	ErrorResponse
410	Gone	ErrorResponse
500	Internal Server Error	ErrorResponse
501	Not Implemented	ErrorResponse

Models

ErrorResponse

Name	Type	Description	Required
code	integer	The HTTP status code.	Yes
error	string	The error message.	Yes
detail	string	The details of the error.	No

TrackingData

Name	Type	Description	Required
title	string	The title of the job.	No
reference	string	External system reference.	No
tags	[string]		No
details	object	Customer-defined JSON structure.	No

DataFetchConfig

Name	Type	Description	Required
url	string	A URL where a file is stored	Yes
auth_headers	[string]	A list of additional headers to be added to the input fetch request when using http or https. This is intended to support authentication or authorization, for example by supplying an OAuth2 bearer token.	No

TranscriptionConfig

Name	Type	Description	Required
language	string	Language model to process the audio input, normally specified as an ISO language code	Yes
output_locale	string	Language locale to be used when generating the transcription output, normally specified as an ISO language code	No
additional_vocab	[object]	List of custom words or phrases that should be recognized. Alternative pronunciations can be specified to aid recognition.	No
punctuation_overrides		Control punctuation settings.	No
diarization	string	Specify whether speaker or channel labels are added to the transcript. The default is none. - none: no speaker or channel labels are added. - speaker: speaker attribution is performed based on acoustic matching; all input channels are mixed into a single stream for processing. - channel: multiple input channels are processed individually and collated into a single transcript. - speaker_change: the output indicates when the speaker in the audio changes. No speaker attribution is performed. This is a faster method than speaker. The reported speaker changes may not agree with speaker. - channel_and_speaker_change: both channel and speaker_change are switched on. The speaker change is indicated if more than one speaker are recorded in one channel.	No
speaker_diarization_config	SpeakerDiarizationConfig	Configuration for speaker diarization. Includes speaker_sensitivity: Range between 0 and 1. A higher sensitivity will increase the likelihood of more unique speakers returning. For example, if you see fewer speakers returned than expected, you can try increasing the sensitivity value or if too many speakers are returned try reducing this value. The default is 0.5.	No
speaker_change_sensitivity	float	Ranges between zero and one. Controls how responsive the system is for potential speaker changes. High value indicates high sensitivity. Defaults to 0.4.	No
channel_diarization_labels	[string]	Transcript labels to use when using collating separate input channels.	No
speaker_diarization_params		(Deprecated, Ignored) Configuration for speaker diarization	No
operating_point	string	Specify whether to use a standard or enhanced model for transription. By default the model used is standard	No
enable_entities	Boolean	Specify whether to enable entity types within JSON output, as well as additional spoken_form and written_form metadata. By default false	No

For the diarization parameter, the following values are valid:

Value	Description
none	no speaker or channel labels are added.
speaker	speaker attribution is performed based on acoustic matching; all input channels are mixed into a single stream for processing.
channel	multiple input channels are processed individually and collated into a single transcript.
speaker_change	the output indicates when the speaker in the audio changes. No speaker attribution is performed. This is a faster method than speaker. The reported speaker changes may not agree with speaker.
channel_and_speaker_change	both channel and speaker_change are switched on. The speaker change is indicated if more than one speaker are recorded in one channel.

SpeakerDiarizationConfig

Additional configuration for the Speaker Diarization feature.

Name	Type	Description	Required
speaker_sensitivity	float	Used for speaker diarization feature. Range between 0 and 1. A higher sensitivity will increase the likelihood of more unique speakers returning. For example, if you see fewer speakers returned than expected, you can try increasing the sensitivity value, or if too many speakers are returned try reducing this value. The default is 0.5.	No

NotificationConfig

Name	Type	Description	Required
url	string	The url to which a notification message will be sent upon completion of the job. The job id and status are added as query parameters, and any combination of the job inputs and outputs can be included by listing them in contents. If contents is empty, the body of the request will be empty. If only one item is listed, it will be sent as the body of the request with Content-Type set to an appropriate value such as application/octet-stream or application/json. If multiple items are listed they will be sent as named file attachments using the multipart content type. If contents is not specified, the transcript item will be sent within the body of the POST request in json-v2 format. If the job was rejected or failed during processing, that will be indicated by the status, and any output items that are not available as a result will be omitted. The body formatting rules will still be followed as if all items were available. The user-agent header is set to Speechmatics-API/2.0, or Speechmatics API V2 in older API versions.	Yes
contents	[string]	Specifies a list of items to be attached to the notification message. When multiple items are requested, they are included as named file attachments.	No
method	string	The method to be used with http and https urls. The default is post.	No
auth_headers	[string]	A list of additional headers to be added to the notification request when using http or https. This is intended to support authentication or authorization, for example by supplying an OAuth2 bearer token.	No

OutputConfig

If you want the transcription output to be in the SubRip Title (SRT) format, and you want to alter the default parameters Speechmatics provides you must provide the output_confiog within the config object

Name	Type	Description	Required
srt_overrides	object	Parameters that override default values of srt conversion. max_line_length: sets maximum count of characters per subtitle line including white space. max_lines: sets maximum count of lines in a subtitle section.	No

JobConfig

JSON object that contains various groups of job configuration parameters. Based on the value of type, a type-specific object such as transcription_config is required to be present to specify all configuration settings or parameters needed to process the job inputs as expected.

If the results of the job are to be forwarded on completion, notification_config can be provided with a list of callbacks to be made; no assumptions should be made about the order in which they will occur.

Customer specific job details or metadata can be supplied in tracking, and this information will be available where possible in the job results and in callbacks.

Name	Type	Description	Required
type	string		Yes
fetch_data	DataFetchConfig		No
fetch_text	DataFetchConfig		No
transcription_config	TranscriptionConfig		No
notification_config	[NotificationConfig]		No
tracking	TrackingData		No
output_config	OutputConfig		No

CreateJobResponse

In the job response you will see balance and cost values returned, but these are not used by the appliance; they are only maintained for backwards compatibility with the legacy V1 Cloud Offering, and should be ignored by clients.

Name	Type	Description	Required
id	string	The unique ID assigned to the job. Keep a record of this for later retrieval of your completed job.	Yes

JobDetails

Document describing a job. JobConfig will be present in JobDetails returned for GET jobs/ request in the Cloud Offering and in Batch Appliance, but it will not be present in JobDetails returned as item in RetrieveJobsResponse in case of Batch Appliance.

Name	Type	Description	Required
created_at	dateTime	The UTC date time the job was created.	Yes
data_name	string	Name of the data file submitted for job.	Yes
duration	integer	The file duration (in seconds). May be missing for fetch URL jobs.	No
id	string	The unique id assigned to the job.	Yes
status	string	The status of the job. running - The job is actively running. done - The job completed successfully. rejected - The job was accepted at first, but later could not be processed by the transcriber. deleted - The user deleted the job. * expired - The system deleted the job. Usually because the job was in the done state for a very long time.	Yes
config	JobConfig		No

RetrieveJobsResponse

Name	Type	Description	Required
jobs	[JobDetails]		Yes

RetrieveJobResponse

Name	Type	Description	Required
job	JobDetails		Yes

DeleteJobResponse

Name	Type	Description	Required
job	JobDetails		Yes

JobInfo

Summary information about an ASR job, to support identification and tracking.

Name	Type	Description	Required
created_at	dateTime	The UTC date time the job was created.	Yes
data_name	string	Name of data file submitted for job.	Yes
duration	integer	The data file audio duration (in seconds).	Yes
id	string	The unique id assigned to the job.	Yes
tracking	TrackingData	customer-supplied data	No

RecognitionMetadata

Summary information about the output from an ASR job, comprising the job type and configuration parameters used when generating the output.

Name	Type	Description	Required
created_at	dateTime	The UTC date time the transcription output was created.	Yes
type	string		Yes
transcription_config	TranscriptionConfig		No
output_config	OutputConfig		No

RecognitionDisplay

Name	Type	Description	Required
direction	string		Yes

RecognitionAlternative

List of possible job output item values, ordered by likelihood.

Name	Type	Description	Required
content	string		Yes
confidence	float		Yes
language	string		Yes
display	RecognitionDisplay		No
speaker	string		No
tags	[string]		No

RecognitionResult

An ASR job output item. The primary item types are word and punctuation. Other item types may be present, for example to provide semantic information of different forms.

Name	Type	Description	Required
channel	string		No
start_time	float		Yes
end_time	float		Yes
entity_class	string	If an entity has been recognised, what type of entity it is. Displayed even if enable_entities is false	Yes
spoken_form	array	For entity results only, the spoken_form is the transcript of the words directly spoken. Only valid if enable_entities is true	No
written_form	array	For entity results only, the written_form is a standardized form of the spoken words. Only valid if enable_entities is true	No
is_eos	boolean	Whether the punctuation mark is an end of sentence character. Only applies to punctuation marks.	No
type	string	New types of items may appear without being requested; unrecognized item types can be ignored. Current types are word, punctuation, speaker_change, and entity	Yes
alternatives	[RecognitionAlternative]		No

RetrieveTranscriptResponse

Name	Type	Description	Required
format	string	Speechmatics JSON transcript format version number.	Yes
job	JobInfo		Yes
metadata	RecognitionMetadata		Yes