Submit a job
POST /batch-processor
This endpoint is used to submit jobs to the batch processor.
A job request includes three components: a preset (preset), an input configuration (inParams), an output configuration (outParams).
The preset defines the type of job to run. This is a required attribute and available presets are: transcript, soap-notes, summarize.
The inParams defines where the inputs for the job are coming from. An existing recordingId or a URL to a video/audio file are the two possible configurations.
The outParams defines how the outputs for the job are configured like the filename of the output.
Depending on the preset type, an optional transformParams object may contain configuration that describes the transformation applied to the data. Currently only the transcript preset supports these transform configuration options.
Request Template (json)
Top-level configuration
| Attribute Name | Type | Description | Default | Example | Required? |
|---|---|---|---|---|---|
preset | string | The type of job to run | null | [transcript, soap-notes, summarize] | Yes |
inParams | object | The input configuration | null | {...} | Yes |
outParams | object | The output configuration | null | {...} | Yes |
transformParams | object | The output configuration | null | {...} | No |
Input configuration (inParams)
| Attribute Name | Type | Description | Default | Example | Required? |
|---|---|---|---|---|---|
sourceType | string | The type of the input media: [uri, recordingId] | null | One of [uri, recordingId] | Yes |
uri | string | The actual URL when using "uri" as sourceType | null | "https://direct-url-to/file.mp4" | When using uri as sourceType |
recordingId | string | The recording ID when using recordingId as sourceType | null | "uuiasdfe-8ba2-4ee6-bd15-003a92c18245" | When using recordingId as sourceType |
language | string | The BCP-47 language-tag of spoken the audio for the transcription | en | "fr" | No |
Note: language in inParams is only supported for the transcript preset. Additional options are available for transcriptions, see below.
Transform configuration (transformParams)
| Attribute Name | Type | Description | Default | Example | Required? |
|---|---|---|---|---|---|
transcript | object | Configuration specific to transcription | null | {...} | No |
Transcription configuration (transformParams.transcript)
The following parameters can be passed in the transformParams.transcript object.
None of the parameters are required. See the documentation for each parameter for more information on their default values.
Note that you can include more parameters using the extra object.
If you pass a value for language both using inParams and this configuration object, the value provided here will be used.
| Name | Type | Description |
|---|---|---|
language | string | See Deepgram's documentation for language |
model | string | See Deepgram's documentation for model |
tier | string | This field is deprecated, use model instead |
profanity_filter | boolean | See Deepgram's documentation for profanity_filter |
punctuate | boolean | See Deepgram's documentation for punctuate |
endpointing | number or boolean | See Deepgram's documentation for endpointing |
redact | boolean or array | See Deepgram's documentation for redact |
extra | object | Specify additional parameters. See Deepgram's documentation for available streaming options |
Language Support
The exact list of supported languages may depend on the model parameter in transcription configuration (see above).
In the default configuration, the supported languages using the language parameter are:
| Language | BCP-47 tag |
|---|---|
| Danish | da |
| Dutch | nl |
| English | en |
| English Australia | en-AU |
| English UK | en-GB |
| English India | en-IN |
| English NewZealand | en-NZ |
| English USA | en-US |
| French | fr |
| French Canada | fr-CA |
| German (DE) | de |
| Hindi | hi |
| Hindi Roman | hi-Latn |
| Indonedian | id |
| Italian | it |
| Japanese | ja |
| Korean | ko |
| Norwegian | no |
| Polish | pl |
| Portuguese | pt |
| Portuguese Brazil | pt-BR |
| Portuguese Portugal | pt-PT |
| Russian | ru |
| Spanish | es |
| Spanish Latin America | es-419 |
| Swedish | sv |
| Turkish | tr |
| Ukrainian | uk |
Output configuration (outParams)
| Attribute Name | Type | Description | Default | Example | Required? |
|---|---|---|---|---|---|
s3config.s3KeyTemplate | string | The filename of the output | null | "my-soap-note" | Yes |
s3config.useReplacement | boolean | whether to use template replacement in s3KeyTemplate | null | true | No |
s3KeyReplacement
The default output path for transcript, soap and summary jobs are domain_name/jobId/transcript/s3Config.s3KeyTemplate domain_name/jobId/soap/s3Config.s3KeyTemplate domain_name/jobId/summary/s3Config.s3KeyTemplate respectively.
s3KeyReplacement with useReplacement can be used to customize the S3 path of the generated output, template replacements are made up of a replacement string with prefixes, suffixes, or both. The currently supported replacements are:
epoch_time: The epoch time in seconds (optional)
domain_name: Your Daily domain (optional)
job_id: The job ID assigned to the job (optional)
The restrictions for defining a recording template are as follows:
The maximum size of the template is 1024 characters
Each replacement parameter should be placed within a curly bracket (e.g., {domain*name})
Only alphanumeric characters (0-9, A-Z, a-z) and ., /, -, * are valid within the template .mp4 is the only valid extension
Examples
Example domain: "myDomain"
Template: myprefix-{domain_name}-{epoch_time}
Resulting transcripts names:
myprefix-myDomain-1675842936274.vtt
myprefix-myDomain-1675842936274.json
myprefix-myDomain-1675842936274.txt