@runware/schemas 0.0.0-20260510231619

package/dist/schemas/requests/models/runware/dia2-2b.json

@@ -0,0 +1,809 @@
+{
+  "$schema": "https://json-schema.org/draft/2019-09/schema",
+  "type": "object",
+  "properties": {
+    "model": {
+      "title": "Model",
+      "description": "Identifier of the model to use for generation.",
+      "type": "string",
+      "const": "runware:dia2@2b"
+    },
+    "inputs": {
+      "title": "Inputs",
+      "description": "The unified payload wrapper for complex media assets dictating image, video or audio inference constraints.",
+      "type": "object",
+      "properties": {
+        "audios": {
+          "title": "Audios",
+          "description": "Voice reference audio samples for speaker cloning (UUID, URL, or Base64). audios[0] = Speaker 1 ([S1]), audios[1] = Speaker 2 ([S2]). Internally transcribed with Whisper for alignment.",
+          "type": "array",
+          "minItems": 1,
+          "items": {
+            "title": "Audio",
+            "description": "Audio input (UUID or URL).",
+            "type": "string",
+            "anyOf": [
+              {
+                "format": "uuid"
+              },
+              {
+                "format": "uri"
+              }
+            ]
+          },
+          "maxItems": 2
+        }
+      },
+      "additionalProperties": false
+    },
+    "negativePrompt": {
+      "title": "Negative Prompt",
+      "description": "Prompt to guide what to exclude from generation. Ignored when guidance is disabled (CFGScale ≤ 1).",
+      "type": "string",
+      "minLength": 2,
+      "maxLength": 3000
+    },
+    "seed": {
+      "title": "Seed",
+      "description": "Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.",
+      "type": "integer",
+      "minimum": 0,
+      "maximum": 2147483647
+    },
+    "CFGScale": {
+      "title": "CFG Scale",
+      "description": "Guidance scale representing how closely the output will resemble the prompt. Higher values produce results more aligned with the prompt.",
+      "type": "number",
+      "multipleOf": 0.01,
+      "minimum": 1,
+      "default": 3,
+      "maximum": 30
+    },
+    "speech": {
+      "title": "Speech",
+      "description": "Settings for speech generation.",
+      "type": "object",
+      "properties": {
+        "text": {
+          "title": "Speech Text",
+          "description": "Text to convert to speech.",
+          "type": "string",
+          "minLength": 1,
+          "maxLength": 3000
+        }
+      },
+      "required": [
+        "text"
+      ],
+      "additionalProperties": false
+    },
+    "settings": {
+      "title": "Settings",
+      "description": "Global inference settings.",
+      "type": "object",
+      "properties": {
+        "temperature": {
+          "type": "number",
+          "title": "Temperature",
+          "description": "Sampling temperature for text token generation. Lower = more coherent, higher = more variety. 0 = greedy decoding.",
+          "default": 0.6,
+          "minimum": 0,
+          "maximum": 5
+        },
+        "audioTemperature": {
+          "type": "number",
+          "title": "Audio Temperature",
+          "description": "Sampling temperature for audio token generation. Lower = cleaner audio, higher = more expressive. 0 = greedy decoding.",
+          "default": 0.8,
+          "minimum": 0,
+          "maximum": 5
+        },
+        "includePrefix": {
+          "type": "boolean",
+          "title": "Include Prefix",
+          "description": "Whether to include prefix audio in the final output or trim it.",
+          "default": false
+        },
+        "topK": {
+          "type": "integer",
+          "title": "Top K",
+          "description": "Top-K sampling for both text and audio tokens. Limits selection to K most likely tokens per step.",
+          "default": 50,
+          "minimum": 1,
+          "maximum": 500
+        }
+      },
+      "additionalProperties": false
+    },
+    "taskType": {
+      "title": "Task Type",
+      "description": "Identifier for the type of task being performed",
+      "type": "string",
+      "const": "audioInference"
+    },
+    "taskUUID": {
+      "title": "Task UUID",
+      "description": "UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.",
+      "type": "string",
+      "format": "uuid"
+    },
+    "webhookURL": {
+      "title": "Webhook URL",
+      "description": "Specifies a webhook URL where JSON responses will be sent via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.",
+      "type": "string",
+      "format": "uri"
+    },
+    "includeCost": {
+      "title": "Include Cost",
+      "description": "Include task cost in the response.",
+      "type": "boolean",
+      "default": false
+    },
+    "numberResults": {
+      "title": "Number of Results",
+      "description": "Number of results to generate. Each result uses a different seed, producing variations of the same parameters.",
+      "type": "integer",
+      "minimum": 1,
+      "default": 1,
+      "maximum": 4
+    },
+    "uploadEndpoint": {
+      "title": "Upload Endpoint",
+      "description": "URL to upload content via HTTP PUT. The request body will contain the raw binary data.",
+      "type": "string",
+      "format": "uri"
+    },
+    "ttl": {
+      "title": "TTL",
+      "description": "Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.",
+      "type": "integer",
+      "minimum": 60
+    },
+    "outputType": {
+      "title": "Output Type",
+      "description": "Audio output type.",
+      "type": "string",
+      "enum": [
+        "URL",
+        "base64Data",
+        "dataURI"
+      ],
+      "default": "URL"
+    },
+    "outputFormat": {
+      "title": "Output Format",
+      "description": "File format for the generated audio.",
+      "type": "string",
+      "enum": [
+        "MP3",
+        "WAV",
+        "FLAC",
+        "OGG"
+      ],
+      "default": "MP3"
+    },
+    "deliveryMethod": {
+      "title": "Delivery Method",
+      "description": "Determines how the API delivers task results.",
+      "type": "string",
+      "oneOf": [
+        {
+          "const": "sync",
+          "description": "Returns complete results directly in the API response."
+        },
+        {
+          "const": "async",
+          "description": "Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse."
+        }
+      ],
+      "default": "sync"
+    },
+    "audioSettings": {
+      "title": "Audio Settings",
+      "description": "Audio encoding settings. Only applicable for lossy formats (MP3, OGG). Available sample rates and valid bitrate ranges depend on the output format and number of channels.",
+      "type": "object",
+      "properties": {
+        "bitrate": {
+          "title": "Bitrate",
+          "description": "Audio bitrate in kbps.",
+          "type": "integer",
+          "minimum": 8
+        },
+        "channels": {
+          "title": "Channels",
+          "description": "Number of audio channels. 1 for mono, 2 for stereo.",
+          "type": "integer",
+          "enum": [
+            1,
+            2
+          ],
+          "default": 2
+        },
+        "sampleRate": {
+          "title": "Sample Rate",
+          "description": "Audio sample rate in Hz.",
+          "type": "integer"
+        }
+      },
+      "additionalProperties": false
+    }
+  },
+  "required": [
+    "speech",
+    "taskType",
+    "taskUUID",
+    "model"
+  ],
+  "allOf": [
+    {
+      "if": {
+        "properties": {
+          "outputFormat": {
+            "enum": [
+              "WAV",
+              "FLAC"
+            ]
+          }
+        },
+        "required": [
+          "outputFormat"
+        ]
+      },
+      "then": {
+        "not": {
+          "required": [
+            "audioSettings"
+          ]
+        }
+      }
+    },
+    {
+      "if": {
+        "properties": {
+          "outputFormat": {
+            "const": "MP3"
+          }
+        },
+        "required": [
+          "outputFormat"
+        ]
+      },
+      "then": {
+        "properties": {
+          "audioSettings": {
+            "properties": {
+              "sampleRate": {
+                "enum": [
+                  8000,
+                  11025,
+                  12000,
+                  16000,
+                  22050,
+                  24000,
+                  32000,
+                  44100,
+                  48000
+                ]
+              }
+            }
+          }
+        },
+        "allOf": [
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "enum": [
+                        8000,
+                        11025,
+                        12000
+                      ]
+                    }
+                  },
+                  "required": [
+                    "sampleRate"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 8,
+                      "maximum": 64
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "enum": [
+                        16000,
+                        22050,
+                        24000
+                      ]
+                    }
+                  },
+                  "required": [
+                    "sampleRate"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 8,
+                      "maximum": 160
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "enum": [
+                        32000,
+                        44100,
+                        48000
+                      ]
+                    }
+                  },
+                  "required": [
+                    "sampleRate"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 32,
+                      "maximum": 320
+                    }
+                  }
+                }
+              }
+            }
+          }
+        ]
+      }
+    },
+    {
+      "if": {
+        "properties": {
+          "outputFormat": {
+            "const": "OGG"
+          }
+        },
+        "required": [
+          "outputFormat"
+        ]
+      },
+      "then": {
+        "properties": {
+          "audioSettings": {
+            "properties": {
+              "sampleRate": {
+                "enum": [
+                  8000,
+                  12000,
+                  16000,
+                  24000,
+                  48000
+                ]
+              }
+            }
+          }
+        },
+        "allOf": [
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 8000
+                    },
+                    "channels": {
+                      "const": 1
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 8,
+                      "maximum": 40
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 8000
+                    },
+                    "channels": {
+                      "const": 2
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 16,
+                      "maximum": 80
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 12000
+                    },
+                    "channels": {
+                      "const": 1
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 16,
+                      "maximum": 48
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 12000
+                    },
+                    "channels": {
+                      "const": 2
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 16,
+                      "maximum": 96
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 16000
+                    },
+                    "channels": {
+                      "const": 1
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 16,
+                      "maximum": 96
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 16000
+                    },
+                    "channels": {
+                      "const": 2
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 24,
+                      "maximum": 192
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 24000
+                    },
+                    "channels": {
+                      "const": 1
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 16,
+                      "maximum": 80
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 24000
+                    },
+                    "channels": {
+                      "const": 2
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 32,
+                      "maximum": 160
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 48000
+                    },
+                    "channels": {
+                      "const": 1
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 32,
+                      "maximum": 224
+                    }
+                  }
+                }
+              }
+            }
+          },
+          {
+            "if": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "sampleRate": {
+                      "const": 48000
+                    },
+                    "channels": {
+                      "const": 2
+                    }
+                  },
+                  "required": [
+                    "sampleRate",
+                    "channels"
+                  ]
+                }
+              },
+              "required": [
+                "audioSettings"
+              ]
+            },
+            "then": {
+              "properties": {
+                "audioSettings": {
+                  "properties": {
+                    "bitrate": {
+                      "minimum": 48,
+                      "maximum": 256
+                    }
+                  }
+                }
+              }
+            }
+          }
+        ]
+      }
+    },
+    {
+      "if": {
+        "properties": {
+          "settings": {
+            "required": [
+              "includePrefix"
+            ]
+          }
+        },
+        "required": [
+          "settings"
+        ]
+      },
+      "then": {
+        "properties": {
+          "inputs": {
+            "required": [
+              "audios"
+            ]
+          }
+        },
+        "required": [
+          "inputs"
+        ]
+      }
+    }
+  ],
+  "additionalProperties": false
+}