> ## Documentation Index > Fetch the complete documentation index at: https://docs.slng.ai/llms.txt > Use this file to discover all available pages before exploring further. # Create batch job > Submit audio for asynchronous transcription. Supports file upload (`multipart/form-data`), URL input, and presigned S3 upload (`application/json`). For the presigned-upload flow, the first call (`mode: "presign"`) returns **200 OK** with an upload URL — the job is not created until step 3. All other submission paths return **202 Accepted** with `status: QUEUED`. ## OpenAPI ````yaml /api-reference/batch/batch.oas.json post /v1/batch/jobs openapi: 3.1.0 info: version: 0.1.0 title: SLNG Batch API description: >- Transcribe audio files asynchronously. Upload a file, monitor the job progress, and download the transcript when it's ready. servers: - url: https://api.batch.slng.ai description: Production - url: https://stageapi.batch.slng.ai description: Staging security: - bearerAuth: [] tags: - name: Speechmatics description: Create and manage asynchronous transcription jobs. paths: /v1/batch/jobs: post: tags: - Speechmatics summary: Create batch job description: >- Submit audio for asynchronous transcription. Supports file upload (`multipart/form-data`), URL input, and presigned S3 upload (`application/json`). For the presigned-upload flow, the first call (`mode: "presign"`) returns **200 OK** with an upload URL — the job is not created until step 3. All other submission paths return **202 Accepted** with `status: QUEUED`. operationId: batchJobsCreate requestBody: required: true content: application/json: schema: oneOf: - title: URL input type: object required: - input_url - transcription_config properties: input_url: type: string format: uri description: >- HTTPS URL of the audio file to transcribe. Must be publicly accessible or a presigned URL. job_id: type: string description: >- Custom alphanumeric identifier for the job. Auto-generated if omitted. pattern: ^[a-zA-Z0-9]+$ model_code: type: string description: Model to use for transcription. default: slng/speechmatics/batch:15.0.0 transcription_config: $ref: '#/components/schemas/TranscriptionConfig' metadata: type: object description: Arbitrary key-value metadata to attach to the job. tracking: $ref: '#/components/schemas/Tracking' output_config: $ref: '#/components/schemas/OutputConfig' - title: 'Presigned upload: request a URL' type: object required: - mode - filename - transcription_config properties: mode: type: string enum: - presign description: Set to `presign` to request a presigned S3 upload URL. filename: type: string description: Name of the file to upload. content_type: type: string description: >- MIME type of the file. Default: `application/octet-stream`. job_id: type: string description: >- Custom alphanumeric identifier for the job. Auto-generated if omitted. pattern: ^[a-zA-Z0-9]+$ transcription_config: $ref: '#/components/schemas/TranscriptionConfig' tracking: $ref: '#/components/schemas/Tracking' output_config: $ref: '#/components/schemas/OutputConfig' - title: 'Presigned upload: create the job' type: object required: - s3_key - transcription_config properties: s3_key: type: string description: S3 key returned from the presign step. job_id: type: string description: Job ID returned from the presign step. pattern: ^[a-zA-Z0-9]+$ model_code: type: string description: Model to use for transcription. transcription_config: $ref: '#/components/schemas/TranscriptionConfig' metadata: type: object description: Arbitrary key-value metadata to attach to the job. tracking: $ref: '#/components/schemas/Tracking' output_config: $ref: '#/components/schemas/OutputConfig' multipart/form-data: schema: type: object required: - file - transcription_config properties: file: type: string format: binary description: >- The audio file to transcribe. Supported formats: `wav`, `mp3`, `flac`, `aac`, `ogg`, `m4a`, `mp4`, `amr`, `mpeg`. job_id: type: string description: >- Custom alphanumeric identifier for the job. Auto-generated if omitted. pattern: ^[a-zA-Z0-9]+$ example: myjob123 metadata: type: string description: >- JSON object with arbitrary key-value metadata to attach to the job. model_code: type: string description: Model to use for transcription. default: slng/speechmatics/batch:15.0.0 transcription_config: $ref: '#/components/schemas/TranscriptionConfig' tracking: $ref: '#/components/schemas/Tracking' output_config: $ref: '#/components/schemas/OutputConfig' encoding: transcription_config: contentType: application/json tracking: contentType: application/json output_config: contentType: application/json metadata: contentType: application/json responses: '200': description: >- **Presign mode only.** Returns a presigned upload URL; the job is not created yet. See the [Batch guide](/batch-guide#method-3-presigned-upload) for the full upload flow. content: application/json: schema: type: object required: - job_id - transcription_config - input_s3_uri - output_s3_prefix - upload properties: job_id: type: string description: Job identifier to reuse in step 3. transcription_config: $ref: '#/components/schemas/TranscriptionConfig' tracking: $ref: '#/components/schemas/Tracking' output_config: $ref: '#/components/schemas/OutputConfig' input_s3_uri: type: string description: '`s3://bucket/key` of the future upload.' output_s3_prefix: type: string description: >- `s3://bucket/outputs//` where transcripts will land. upload: type: object required: - url - s3_key - s3_uri - expires_in - headers properties: url: type: string format: uri description: Presigned `PUT` URL. s3_key: type: string description: S3 key the URL writes to. Pass as `s3_key` in step 3. s3_uri: type: string description: Same key, full `s3://` URI. expires_in: type: integer description: URL lifetime in seconds (30 min). headers: type: object additionalProperties: type: string description: >- Headers the upload request must send (e.g. `Content-Type`). example: job_id: job8f3a1b2c4d5e6f7890abcdef12345678 transcription_config: language: en operating_point: standard diarization: none input_s3_uri: s3://slng-production-batch-input/inputs/job8f3a.../audio.wav output_s3_prefix: s3://slng-production-batch-output/outputs/job8f3a.../ upload: url: >- https://slng-production-batch-input.s3.amazonaws.com/inputs/job8f3a.../audio.wav?X-Amz-Algorithm=... s3_key: inputs/job8f3a.../audio.wav s3_uri: s3://slng-production-batch-input/inputs/job8f3a.../audio.wav expires_in: 1800 headers: Content-Type: application/octet-stream '202': description: >- Job accepted and queued for processing (multipart upload, URL input, or presigned-upload step 3). content: application/json: schema: $ref: '#/components/schemas/JobAcknowledgement' example: job_id: myjob123 status: QUEUED '400': description: >- Validation error (bad request body, unsupported `model_code`, invalid `job_id`, missing/invalid `transcription_config`, invalid `input_url`, etc.). content: application/json: schema: $ref: '#/components/schemas/Error' example: error: job_id must be alphanumeric '403': $ref: '#/components/responses/Forbidden' '413': description: File too large. Maximum size is 1 GB. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: File too large. Maximum size is 1 GB. '500': $ref: '#/components/responses/InternalError' components: schemas: TranscriptionConfig: type: object description: Transcription settings. properties: language: type: string description: Language code for the audio. default: en example: en operating_point: type: string description: Accuracy level. enum: - standard - enhanced default: standard diarization: type: string description: Speaker separation mode. enum: - none - speaker - channel default: none domain: type: string description: Domain-specific language model to use. output_locale: type: string description: Locale for formatting output (e.g. dates, numbers). enable_entities: type: boolean description: Detect and format entities such as dates, times, and addresses. additional_vocab: type: array description: Custom words or phrases to improve recognition accuracy. items: type: object properties: content: type: string sounds_like: type: array items: type: string Tracking: type: object description: Metadata for organizing and filtering jobs. properties: title: type: string description: Human-readable job title. example: Meeting recording reference: type: string description: External reference identifier. example: ref-123 tags: type: array description: Labels for categorizing the job. items: type: string example: - meeting - q1-review details: type: string description: Free-text notes about the job. example: Q1 planning session OutputConfig: type: object description: Output format settings. properties: srt_overrides: type: object description: SRT subtitle formatting options. properties: max_line_length: type: integer description: Maximum characters per subtitle line. example: 80 max_lines: type: integer description: Maximum lines per subtitle block. example: 2 JobAcknowledgement: type: object description: >- Minimal acknowledgement returned when a job is queued or marked for deletion. Poll `GET /v1/batch/jobs/{jobId}` for the full record. required: - job_id - status properties: job_id: type: string description: Alphanumeric job identifier. status: type: string description: Initial status. `QUEUED` on creation; `DELETED` on deletion. Error: type: object description: >- Error envelope. A short, user-presentable string only — no internal codes, stack traces, or upstream raw text. required: - error additionalProperties: false properties: error: type: string responses: Forbidden: description: Missing or invalid API key. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: API key id not found InternalError: description: Internal server error. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: Internal server error securitySchemes: bearerAuth: type: http scheme: bearer description: API key from the SLNG dashboard. ````