JobConfig

{
  "$schema": "http://json-schema.org/draft-07/schema#",
  "$id": "https://www.speechmatics.com/schema/job-config",
  "title": "JobConfig",
  "description": "Configuration object for a Speechmatics batch transcription job.",
  "type": "object",
  "required": ["type"],
  "properties": {
    "type": {
      "type": "string",
      "enum": ["transcription"],
      "description": "The type of job. Currently only 'transcription' is supported."
    },
    "transcription_config": {
      "$ref": "#/definitions/TranscriptionConfig"
    },
    "translation_config": {
      "$ref": "#/definitions/TranslationConfig"
    },
    "language_identification_config": {
      "$ref": "#/definitions/LanguageIdentificationConfig"
    },
    "summarization_config": {
      "$ref": "#/definitions/SummarizationConfig"
    },
    "sentiment_analysis_config": {
      "type": "object",
      "description": "Enable sentiment analysis on the transcript."
    },
    "topic_detection_config": {
      "$ref": "#/definitions/TopicDetectionConfig"
    },
    "auto_chapters_config": {
      "type": "object",
      "description": "Enable automatic chapter detection."
    },
    "audio_events_config": {
      "$ref": "#/definitions/AudioEventsConfig"
    },
    "output_config": {
      "$ref": "#/definitions/OutputConfig"
    },
    "notification_config": {
      "type": "array",
      "items": {
        "$ref": "#/definitions/NotificationConfig"
      },
      "description": "Webhook callbacks to invoke on job completion."
    },
    "tracking": {
      "$ref": "#/definitions/TrackingData"
    },
    "fetch_data": {
      "$ref": "#/definitions/DataFetchConfig"
    },
    "fetch_text": {
      "$ref": "#/definitions/DataFetchConfig"
    }
  },
  "definitions": {
    "TranscriptionConfig": {
      "type": "object",
      "required": ["language"],
      "properties": {
        "language": {
          "type": "string",
          "description": "ISO language code for the audio (e.g. 'en', 'fr', 'de', 'auto')."
        },
        "domain": {
          "type": "string",
          "description": "Specialized domain (e.g. 'finance', 'medical')."
        },
        "output_locale": {
          "type": "string",
          "description": "ISO locale for the output text."
        },
        "model": {
          "type": "string",
          "enum": ["standard", "enhanced"],
          "description": "Model quality selector."
        },
        "additional_vocab": {
          "type": "array",
          "items": {
            "type": "object",
            "required": ["content"],
            "properties": {
              "content": {
                "type": "string"
              },
              "sounds_like": {
                "type": "array",
                "items": {
                  "type": "string"
                }
              }
            }
          },
          "description": "Custom vocabulary words/phrases."
        },
        "diarization": {
          "type": "string",
          "enum": ["none", "speaker", "channel"],
          "description": "Speaker attribution mode."
        },
        "channel_diarization_labels": {
          "type": "array",
          "items": {
            "type": "string",
            "pattern": "^[A-Za-z0-9._]+$"
          }
        },
        "enable_entities": {
          "type": "boolean",
          "description": "Include entity annotations in results."
        },
        "punctuation_overrides": {
          "type": "object",
          "properties": {
            "sensitivity": {
              "type": "number",
              "minimum": 0,
              "maximum": 1
            },
            "permitted_marks": {
              "type": "array",
              "items": {
                "type": "string"
              }
            }
          }
        },
        "speaker_diarization_config": {
          "type": "object",
          "properties": {
            "prefer_current_speaker": {"type": "boolean"},
            "speaker_sensitivity": {"type": "number", "minimum": 0, "maximum": 1},
            "get_speakers": {"type": "boolean"}
          }
        },
        "transcript_filtering_config": {
          "type": "object",
          "properties": {
            "remove_disfluencies": {"type": "boolean"},
            "replacements": {
              "type": "array",
              "items": {
                "type": "object",
                "required": ["from", "to"],
                "properties": {
                  "from": {"type": "string"},
                  "to": {"type": "string"}
                },
                "additionalProperties": false
              }
            }
          }
        },
        "audio_filtering_config": {
          "type": "object",
          "properties": {
            "volume_threshold": {
              "type": "number",
              "minimum": 0,
              "maximum": 100
            }
          }
        }
      }
    },
    "TranslationConfig": {
      "type": "object",
      "required": ["target_languages"],
      "properties": {
        "target_languages": {
          "type": "array",
          "maxItems": 5,
          "items": {
            "type": "string"
          },
          "description": "ISO language codes for translation targets."
        }
      }
    },
    "LanguageIdentificationConfig": {
      "type": "object",
      "properties": {
        "expected_languages": {
          "type": "array",
          "items": {"type": "string"}
        },
        "low_confidence_action": {
          "type": "string",
          "enum": ["allow", "reject", "use_default_language"]
        },
        "default_language": {
          "type": "string"
        }
      }
    },
    "SummarizationConfig": {
      "type": "object",
      "properties": {
        "content_type": {
          "type": "string",
          "enum": ["auto", "informative", "conversational"]
        },
        "summary_length": {
          "type": "string",
          "enum": ["brief", "detailed"],
          "default": "brief"
        },
        "summary_type": {
          "type": "string",
          "enum": ["paragraphs", "bullets"]
        }
      }
    },
    "TopicDetectionConfig": {
      "type": "object",
      "properties": {
        "topics": {
          "type": "array",
          "items": {"type": "string"}
        }
      }
    },
    "AudioEventsConfig": {
      "type": "object",
      "properties": {
        "types": {
          "type": "array",
          "items": {"type": "string"}
        }
      }
    },
    "OutputConfig": {
      "type": "object",
      "properties": {
        "srt_overrides": {
          "type": "object",
          "properties": {
            "max_line_length": {"type": "integer"},
            "max_lines": {"type": "integer"}
          }
        }
      }
    },
    "NotificationConfig": {
      "type": "object",
      "required": ["url"],
      "properties": {
        "url": {"type": "string", "format": "uri"},
        "contents": {
          "type": "array",
          "items": {
            "type": "string",
            "enum": ["jobinfo", "transcript", "transcript.json-v2", "transcript.txt", "transcript.srt", "data", "text"]
          }
        },
        "method": {
          "type": "string",
          "enum": ["post", "put"]
        },
        "auth_headers": {
          "type": "array",
          "items": {"type": "string"}
        }
      }
    },
    "TrackingData": {
      "type": "object",
      "properties": {
        "title": {"type": "string"},
        "reference": {"type": "string"},
        "tags": {
          "type": "array",
          "items": {"type": "string"}
        },
        "details": {
          "type": "object",
          "additionalProperties": true
        }
      }
    },
    "DataFetchConfig": {
      "type": "object",
      "required": ["url"],
      "properties": {
        "url": {"type": "string", "format": "uri"},
        "auth_headers": {
          "type": "array",
          "items": {"type": "string"}
        }
      }
    }
  }
}