---
title: "Create a job for Salad Transcription API"
method: POST
path: "/organizations/{organization_name}/inference-endpoints/transcribe/jobs"
tags: ["inference_endpoints"]
---

# Create a job for Salad Transcription API

`POST /organizations/{organization_name}/inference-endpoints/transcribe/jobs`

Creates a job for Salad Transcription API

## Request body

- CreateSaladCloudTranscriptionAPIJob — Create a job for Salad Transcription API
  - `input` SaladCloudTranscriptionAPIInput, required
    - `url` string, uri, required — URL of the audio file to process. Must be a downloadable link.
    - `language_code` 'en' | 'zh' | 'de' | 'es' | 'ru' | 'ko' | 'fr' | 'ja' | 'pt' | 'tr' | 'pl' | 'ca' | 'nl' | 'ar' | 'sv' | 'it' | 'id' | 'hi' | 'fi' | 'vi' | 'he' | 'uk' | 'el' | 'ms' | 'cs' | 'ro' | 'da' | 'hu' | 'ta' | 'no' | 'th' | 'ur' | 'hr' | 'bg' | 'lt' | 'la' | 'mi' | 'ml' | 'cy' | 'sk' | 'te' | 'fa' | 'lv' | 'bn' | 'sr' | 'az' | 'sl' | 'kn' | 'et' | 'mk' | 'br' | 'eu' | 'is' | 'hy' | 'ne' | 'mn' | 'bs' | 'kk' | 'sq' | 'sw' | 'gl' | 'mr' | 'pa' | 'si' | 'km' | 'sn' | 'yo' | 'so' | 'af' | 'oc' | 'ka' | 'be' | 'tg' | 'sd' | 'gu' | 'am' | 'yi' | 'lo' | 'uz' | 'fo' | 'ht' | 'ps' | 'tk' | 'nn' | 'mt' | 'sa' | 'lb' | 'my' | 'bo' | 'tl' | 'mg' | 'as' | 'tt' | 'haw' | 'ln' | 'ha' | 'ba' | 'jw' | 'su' | 'yue' — Language code for the transcription. Defaults to 'en' (English).
    - `return_as_file` boolean — Whether to return the result as a file.
    - `sentence_level_timestamps` boolean — Return sentence-level timestamps in the transcription.
    - `word_level_timestamps` boolean — Return word-level timestamps in the transcription.
    - `diarization` boolean — Perform speaker diarization (separate speakers in multi-speaker audio).
    - `sentence_diarization` boolean — Perform sentence-level speaker diarization.
    - `srt` boolean — Generate an SRT file (SubRip Subtitle format).
    - `translate` string — Translate the transcription to English (set to 'to eng' to enable translation).
    - `custom_prompt` string — Custom prompt for LLM-driven tasks such as summarization or classification.
    - `summarize` integer — Summarize the transcription. Integer specifies the word limit.
    - `llm_translation` string — List of languages for LLM-based translation, e.g., 'german, italian, french, english, portuguese, hindi, spanish, thai'.
    - `srt_translation` string — List of languages for SRT translation, e.g., 'german, italian, french, english, portuguese, hindi, spanish, thai'.
    - `custom_vocabulary` string — Custom vocabulary to improve transcription accuracy for specific terms.
    - `classification_labels` string — List of labels to use for sentence classification, e.g., 'INTERVIEW, MEETING, CALL'.
    - `overall_classification` boolean — Perform overall classification of the transcription.
    - `overall_sentiment_analysis` boolean — Perform overall sentiment analysis on the transcription.
  - `metadata` object, nullable
  - `webhook` string, url, nullable

## Response `201`

Created

- SaladCloudTranscriptionAPIJob — Job input schema for Salad Transcription API
  - `id` string, uuid, required
  - `input` SaladCloudTranscriptionAPIInput, required
    - `url` string, uri, required — URL of the audio file to process. Must be a downloadable link.
    - `language_code` 'en' | 'zh' | 'de' | 'es' | 'ru' | 'ko' | 'fr' | 'ja' | 'pt' | 'tr' | 'pl' | 'ca' | 'nl' | 'ar' | 'sv' | 'it' | 'id' | 'hi' | 'fi' | 'vi' | 'he' | 'uk' | 'el' | 'ms' | 'cs' | 'ro' | 'da' | 'hu' | 'ta' | 'no' | 'th' | 'ur' | 'hr' | 'bg' | 'lt' | 'la' | 'mi' | 'ml' | 'cy' | 'sk' | 'te' | 'fa' | 'lv' | 'bn' | 'sr' | 'az' | 'sl' | 'kn' | 'et' | 'mk' | 'br' | 'eu' | 'is' | 'hy' | 'ne' | 'mn' | 'bs' | 'kk' | 'sq' | 'sw' | 'gl' | 'mr' | 'pa' | 'si' | 'km' | 'sn' | 'yo' | 'so' | 'af' | 'oc' | 'ka' | 'be' | 'tg' | 'sd' | 'gu' | 'am' | 'yi' | 'lo' | 'uz' | 'fo' | 'ht' | 'ps' | 'tk' | 'nn' | 'mt' | 'sa' | 'lb' | 'my' | 'bo' | 'tl' | 'mg' | 'as' | 'tt' | 'haw' | 'ln' | 'ha' | 'ba' | 'jw' | 'su' | 'yue' — Language code for the transcription. Defaults to 'en' (English).
    - `return_as_file` boolean — Whether to return the result as a file.
    - `sentence_level_timestamps` boolean — Return sentence-level timestamps in the transcription.
    - `word_level_timestamps` boolean — Return word-level timestamps in the transcription.
    - `diarization` boolean — Perform speaker diarization (separate speakers in multi-speaker audio).
    - `sentence_diarization` boolean — Perform sentence-level speaker diarization.
    - `srt` boolean — Generate an SRT file (SubRip Subtitle format).
    - `translate` string — Translate the transcription to English (set to 'to eng' to enable translation).
    - `custom_prompt` string — Custom prompt for LLM-driven tasks such as summarization or classification.
    - `summarize` integer — Summarize the transcription. Integer specifies the word limit.
    - `llm_translation` string — List of languages for LLM-based translation, e.g., 'german, italian, french, english, portuguese, hindi, spanish, thai'.
    - `srt_translation` string — List of languages for SRT translation, e.g., 'german, italian, french, english, portuguese, hindi, spanish, thai'.
    - `custom_vocabulary` string — Custom vocabulary to improve transcription accuracy for specific terms.
    - `classification_labels` string — List of labels to use for sentence classification, e.g., 'INTERVIEW, MEETING, CALL'.
    - `overall_classification` boolean — Perform overall classification of the transcription.
    - `overall_sentiment_analysis` boolean — Perform overall sentiment analysis on the transcription.
  - `inference_endpoint_name` string, required — The inference endpoint name
  - `metadata` object, nullable
  - `webhook` string, url, nullable
  - `status` 'pending' | 'running' | 'succeeded' | 'cancelled' | 'failed', required
  - `events` InferenceEndpointJobEvent[], required
    - `action` 'created' | 'started' | 'succeeded' | 'cancelled' | 'failed', required
    - `time` string, date-time, required
  - `organization_name` string, required — The organization name
  - `output` SaladCloudTranscriptionAPIOutput
    - `sentence_level_timestamps` object[] — List of sentences with timestamps, speaker identification, sentiment analysis, and classification.
      - `text` string, required — Transcribed sentence.
      - `timestamp` number[] — Start and end times of the sentence in seconds.
      - `start` number, float, required — Start time of the sentence in seconds.
      - `end` number, float, required — End time of the sentence in seconds.
      - `speaker` string, required — Identified speaker label.
      - `sentiment` object[] — Sentiment analysis results for the sentence.
        - `label` string — Sentiment label.
        - `score` number, float — Confidence score for the sentiment label.
      - `classification` object — Classification labels and their confidence scores for the sentence.
        - `label` string[] — List of classification labels.
        - `scores` number[] — Confidence scores corresponding to each classification label.
    - `word_segments` object[] — List of words with timestamps, confidence scores, and speaker identification.
      - `word` string, required — Transcribed word.
      - `start` number, float, required — Start time of the word in seconds.
      - `end` number, float, required — End time of the word in seconds.
      - `score` number, float, required — Confidence score for the transcribed word.
      - `speaker` string, required — Identified speaker label.
    - `srt_content` string — Generated SRT (SubRip Subtitle) content.
    - `summary` string — Summarized version of the transcription.
    - `llm_translations` object — LLM-generated translations of the transcription.
    - `srt_translation` object — Translations of the SRT content in various languages.
    - `llm_result` string — Main idea or key takeaway generated by the LLM.
    - `overall_classification` string — Overall classification labels for the transcription.
    - `overall_sentiment` string — Overall sentiment of the transcription.
    - `text` string, required — Full transcribed text from the audio.
    - `duration` number, float, required — Duration of the audio file in hours.
    - `processing_time` number, float, required — Time taken to process the transcription in seconds.
  - `create_time` string, date-time, required
  - `update_time` string, date-time, required

## Other responses

- `400` — Bad Request
- `404` — Not Found
- `429` — Too Many Requests
- `default` — Unknown Error

---

[API](https://skmtc.net/salad/apis/saladcloud-api.md) · [All operations](https://skmtc.net/salad/apis/saladcloud-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/salad/saladcloud-api/versions/6ee34c3732f3/schema)