@civitai/client 0.2.0-beta.31 → 0.2.0-beta.32

package/dist/generated/types.gen.d.ts

@@ -1,3 +1,16 @@
+/**
+ * Represents AI vs real image recognition results.
+ */
+export type AiRecognitionResult = {
+  /**
+   * Classification label: "ai" or "real".
+   */
+  label: string;
+  /**
+   * Confidence score for the classification (0.0 to 1.0).
+   */
+  confidence: number;
+};
 /**
  * Base input for AI Toolkit training across all ecosystems
  */
@@ -75,6 +88,78 @@ export type AiToolkitTrainingInput = TrainingInput & {
 } & {
   engine: 'ai-toolkit';
 };
+/**
+ * Input parameters for ACE Step 1.5 audio generation workflow step
+ */
+export type AceStepAudioInput = {
+  musicDescription: string;
+  lyrics: string;
+  /**
+   * Random seed for reproducible generation
+   */
+  seed?: number;
+  /**
+   * Duration in seconds (1-190)
+   */
+  duration?: number;
+  /**
+   * Beats per minute (40-200)
+   */
+  bpm?: number;
+  /**
+   * Time signature (e.g., "4" for 4/4 time)
+   */
+  timeSignature?: string;
+  /**
+   * Language code (e.g., "en", "zh", "ja", "ko")
+   */
+  language?: string;
+  /**
+   * Musical key (e.g., "C major", "E minor")
+   */
+  key?: string;
+  /**
+   * Weight for instrumental elements (0.0-1.0)
+   */
+  instrumentalWeight?: number;
+  /**
+   * Weight for vocal elements (0.0-1.0)
+   */
+  vocalWeight?: number;
+  /**
+   * Optional model override (uses default ACE Step 1.5 turbo if not specified)
+   */
+  model?: string | null;
+};
+/**
+ * Output from ACE Step 1.5 audio generation workflow step
+ */
+export type AceStepAudioOutput = {
+  audioBlob: AudioBlob;
+};
+/**
+ * Workflow step for generating music using ACE Step 1.5.
+ * Produces full songs from text descriptions and structured lyrics.
+ */
+export type AceStepAudioStep = WorkflowStep & {
+  $type: 'aceStepAudio';
+} & {
+  input: AceStepAudioInput;
+  output?: AceStepAudioOutput;
+} & {
+  $type: 'aceStepAudio';
+};
+/**
+ * Workflow step for generating music using ACE Step 1.5.
+ * Produces full songs from text descriptions and structured lyrics.
+ */
+export type AceStepAudioStepTemplate = WorkflowStepTemplate & {
+  $type: 'aceStepAudio';
+} & {
+  input: AceStepAudioInput;
+} & {
+  $type: 'aceStepAudio';
+};
 export type AgeClassificationInput = {
   /**
    * An optional model to use for age classification. If not provided, the default model will determined by the worker
@@ -86,10 +171,51 @@ export type AgeClassificationInput = {
   mediaUrl: string;
 };
 export type AgeClassificationOutput = {
+  /**
+   * Age classification results per image
+   */
   labels: {
     [key: string]: Array<AgeClassifierLabel>;
   };
+  /**
+   * Whether any minor was detected in the dataset
+   */
   hasMinor: boolean;
+  /**
+   * Total number of age predictions made
+   */
+  numPredictions?: number;
+  /**
+   * Number of flagged predictions (minors)
+   */
+  numFlagged?: number;
+  /**
+   * Face recognition results per image (if available)
+   */
+  faceRecognitionFaces?: {
+    [key: string]: Array<FaceDetectionInfo>;
+  } | null;
+  /**
+   * Total number of faces detected across all images
+   */
+  faceRecognitionNumFaces?: number | null;
+  /**
+   * Estimated number of unique people in the dataset (based on similarity clustering)
+   */
+  faceRecognitionNumUniquePeople?: number | null;
+  /**
+   * Whether face recognition data is available for this result
+   */
+  faceRecognitionAvailable?: boolean;
+};
+/**
+ * Represents age classification results for media content.
+ */
+export type AgeClassificationResult = {
+  /**
+   * Array of detected people with age classifications.
+   */
+  detections: Array<AgeDetection>;
 };
 /**
  * Age classification
@@ -117,6 +243,35 @@ export type AgeClassifierLabel = {
   isMinor: boolean;
   boundingBox: Array<number>;
 };
+/**
+ * Represents a single age detection in an image.
+ */
+export type AgeDetection = {
+  /**
+   * The detector type used (e.g., "yolo", "mediapipe", "ensemble").
+   */
+  detectorType: string;
+  boundingBox: BoundingBox;
+  /**
+   * The age label (e.g., "adult", "child", "teen").
+   */
+  ageLabel: string;
+  /**
+   * Confidence score for the age classification (0.0 to 1.0).
+   */
+  confidence: number;
+  /**
+   * Whether the detected person is classified as a minor.
+   */
+  isMinor: boolean;
+  /**
+   * Full probability distribution across all age labels.
+   * Keys are age labels (e.g., "Child 0-12", "Adult 21-44"), values are confidence scores (0.0 to 1.0).
+   */
+  topK?: {
+    [key: string]: number;
+  } | null;
+};
 export declare const AnimalPoseBboxDetector: {
   readonly YOLOX_L_TORCHSCRIPT_PT: 'yolox_l.torchscript.pt';
   readonly YOLOX_L_ONNX: 'yolox_l.onnx';
@@ -131,6 +286,19 @@ export declare const AnimalPoseEstimator: {
   readonly RTMPOSE_M_AP10K_256_ONNX: 'rtmpose-m_ap10k_256.onnx';
 };
 export type AnimalPoseEstimator = (typeof AnimalPoseEstimator)[keyof typeof AnimalPoseEstimator];
+/**
+ * Represents anime vs real image recognition results.
+ */
+export type AnimeRecognitionResult = {
+  /**
+   * Classification label: "anime" or "real".
+   */
+  label: string;
+  /**
+   * Confidence score for the classification (0.0 to 1.0).
+   */
+  confidence: number;
+};
 export declare const AnylineMergeWith: {
   readonly LINEART_STANDARD: 'lineart_standard';
   readonly LINEART_REALISTIC: 'lineart_realistic';
@@ -141,7 +309,9 @@ export type AnylineMergeWith = (typeof AnylineMergeWith)[keyof typeof AnylineMer
 /**
  * An assistant message representing a prior response.
  */
-export type AssistantMessage = {
+export type AssistantMessage = ChatCompletionMessage & {
+  role: 'assistant';
+} & {
   /**
    * The assistant message content (text only).
    */
@@ -154,6 +324,18 @@ export type AssistantMessage = {
    * Optional refusal message if the model refused to respond.
    */
   refusal?: string | null;
+} & {
+  role: 'assistant';
+};
+export type AudioBlob = Blob & {
+  type: 'audio';
+} & {
+  /**
+   * Duration of the audio in seconds
+   */
+  duration?: number | null;
+} & {
+  type: 'audio';
 };
 export type BatchOcrSafetyClassificationInput = {
   mediaUrls: Array<string>;
@@ -197,6 +379,65 @@ export type Blob = {
    */
   blockedReason?: string | null;
 };
+/**
+ * A rectangular region defined by pixel coordinates.
+ */
+export type BlurRegion = {
+  /**
+   * Left edge X coordinate in pixels.
+   */
+  x1: number;
+  /**
+   * Top edge Y coordinate in pixels.
+   */
+  y1: number;
+  /**
+   * Right edge X coordinate in pixels.
+   */
+  x2: number;
+  /**
+   * Bottom edge Y coordinate in pixels.
+   */
+  y2: number;
+};
+/**
+ * Determines how regions are applied to the blur operation.
+ */
+export declare const BlurRegionMode: {
+  readonly INCLUDE: 'include';
+  readonly EXCLUDE: 'exclude';
+};
+/**
+ * Determines how regions are applied to the blur operation.
+ */
+export type BlurRegionMode = (typeof BlurRegionMode)[keyof typeof BlurRegionMode];
+/**
+ * Applies a Gaussian blur with optional region-based masking.
+ */
+export type BlurTransform = ImageTransform & {
+  type: 'blur';
+} & {
+  /**
+   * The blur intensity (1-100).
+   */
+  blur: number;
+  mode: BlurRegionMode;
+  /**
+   * Rectangular regions that define the blur mask.
+   */
+  regions?: Array<BlurRegion>;
+} & {
+  type: 'blur';
+};
+/**
+ * Represents a bounding box with coordinates.
+ */
+export type BoundingBox = {
+  x1: number;
+  y1: number;
+  x2: number;
+  y2: number;
+};
 export declare const BuzzClientAccount: {
   readonly YELLOW: 'yellow';
   readonly BLUE: 'blue';
@@ -216,12 +457,39 @@ export type ChatCompletionChoice = {
   /**
    * The reason the model stopped generating.
    */
-  finishReason: string;
+  finishReason?: string | null;
   /**
    * Log probability information (if requested).
    */
   logprobs?: unknown;
 };
+/**
+ * Base type for message content parts.
+ * Supports both camelCase (imageUrl) and snake_case (image_url) type discriminators via ContentPartJsonConverter.
+ */
+export type ChatCompletionContentPart = {
+  type?: string;
+  /**
+   * The text content.
+   */
+  text?: string | null;
+  imageUrl?: ChatCompletionImageUrl;
+};
+/**
+ * Image URL details matching OpenAI API spec.
+ */
+export type ChatCompletionImageUrl = {
+  /**
+   * The image source (can be a URL, base64 data URI, or raw base64).
+   * After processing, this will contain the blob URL.
+   */
+  url: string;
+  /**
+   * The detail level for processing the image.
+   * "low", "high", or "auto" (default).
+   */
+  detail?: string | null;
+};
 /**
  * Input for a chat completion step, compatible with OpenAI Chat Completions API.
  */
@@ -277,7 +545,7 @@ export type ChatCompletionInput = {
  * Uses ChatCompletionMessageJsonConverter to handle polymorphism and user message content flexibility.
  */
 export type ChatCompletionMessage = {
-  [key: string]: never;
+  role: string;
 };
 /**
  * Output from a chat completion step.
@@ -527,6 +795,47 @@ export type CursedArrayOfTelemetryCursorAndWorkflow = {
   next: string;
   items: Array<Workflow>;
 };
+export type CustomTextToSpeechInput = TextToSpeechInput & {
+  engine: 'custom';
+} & {
+  /**
+   * Reference audio URL for Base voice-cloning mode.
+   * Accepts AIR URNs (existing resources) or HTTP(S) URLs.
+   */
+  refAudioUrl?: string | null;
+  /**
+   * Transcript of the reference audio.
+   * Required for Base mode unless XVectorOnlyMode is true.
+   */
+  refText?: string | null;
+  /**
+   * If true, uses only speaker embedding for Base mode (ref_text not required).
+   */
+  xVectorOnlyMode?: boolean;
+  /**
+   * Built-in speaker name for CustomVoice mode.
+   */
+  speaker?:
+    | 'aiden'
+    | 'dylan'
+    | 'eric'
+    | 'ono_anna'
+    | 'ryan'
+    | 'serena'
+    | 'sohee'
+    | 'uncle_fu'
+    | 'vivian';
+  /**
+   * Optional style instruction for CustomVoice mode.
+   */
+  instruct?: string | null;
+  /**
+   * Optional generation cap for max tokens.
+   */
+  maxNewTokens?: number | null;
+} & {
+  engine: 'custom';
+};
 export declare const DensePoseColormap: {
   readonly 'VIRIDIS (_MAGIC_ANIMATE)': 'Viridis (MagicAnimate)';
   readonly 'PARULA (_CIVIT_AI)': 'Parula (CivitAI)';
@@ -627,6 +936,56 @@ export type EpochResult = {
    */
   blobUrl: string;
 };
+/**
+ * Represents a single face detection with embeddings and landmarks.
+ */
+export type FaceDetection = {
+  boundingBox: BoundingBox;
+  /**
+   * Facial landmarks (e.g., eyes, nose, mouth positions).
+   * Dictionary keys are landmark names, values are (x, y) coordinates.
+   */
+  landmarks?: {
+    [key: string]: ValueTupleOfDoubleAndDouble;
+  } | null;
+  /**
+   * Face embedding vector for similarity comparison.
+   */
+  embedding?: Array<number> | null;
+};
+/**
+ * Information about a detected face in an image
+ */
+export type FaceDetectionInfo = {
+  /**
+   * Bounding box coordinates [x1, y1, x2, y2]
+   */
+  boundingBox: Array<number>;
+  /**
+   * Facial landmarks (e.g., eyes, nose, mouth positions)
+   */
+  landmarks?: {
+    [key: string]: Array<number>;
+  } | null;
+  /**
+   * Whether this face has an embedding (for similarity comparison)
+   */
+  hasEmbedding?: boolean;
+};
+/**
+ * Represents face recognition results for media content.
+ */
+export type FaceRecognitionResult = {
+  /**
+   * Array of detected faces with embeddings and landmarks.
+   */
+  faces: Array<FaceDetection>;
+  /**
+   * Pairwise cosine similarity matrix between detected faces.
+   * Matrix[i][j] represents similarity between face i and face j.
+   */
+  similarityMatrix?: Array<Array<number>> | null;
+};
 export declare const FileFormat: {
   readonly UNKNOWN: 'unknown';
   readonly SAFE_TENSOR: 'safeTensor';
@@ -1099,6 +1458,12 @@ export type ImageJobNetworkParams = {
  */
 export type ImageOutputFormat = {
   format: string;
+  /**
+   * When true, metadata such as EXIF data will be stripped from the output image. Defaults to false.
+   * Note that some formats like JPEG may still include minimal metadata even when this is false, due to encoder behavior.
+   * Setting this to true will attempt to remove all metadata, but results may vary by format and encoder implementation.
+   */
+  hideMetadata?: boolean;
 };
 export declare const ImageResouceTrainingModerationStatus: {
   readonly EVALUATING: 'evaluating';
@@ -1237,7 +1602,14 @@ export type ImageUpscalerInput = {
    * Either A URL, A DataURL or a Base64 string
    */
   image: string;
-  scaleFactor?: number;
+  /**
+   * The upscaler model to use (AIR URN format).
+   */
+  model?: string | null;
+  /**
+   * Number of times to repeat the upscale operation (1-3). Each repeat doubles the resolution.
+   */
+  numberOfRepeats?: number;
 };
 export type ImageUpscalerOutput = {
   blob: ImageBlob;
@@ -1366,6 +1738,51 @@ export declare const KlingModel: {
   readonly V2_5_TURBO: 'v2.5-turbo';
 };
 export type KlingModel = (typeof KlingModel)[keyof typeof KlingModel];
+export type KlingV3ElementInput = {
+  /**
+   * Either A URL, A DataURL or a Base64 string
+   */
+  frontalImage?: string | null;
+  referenceImages?: Array<string>;
+  videoUrl?: string | null;
+};
+export type KlingV3MultiPrompt = {
+  prompt: string;
+  duration?: number;
+};
+export declare const KlingV3Operation: {
+  readonly TEXT_TO_VIDEO: 'text-to-video';
+  readonly IMAGE_TO_VIDEO: 'image-to-video';
+  readonly REFERENCE_TO_VIDEO: 'reference-to-video';
+  readonly VIDEO_TO_VIDEO_EDIT: 'video-to-video-edit';
+  readonly VIDEO_TO_VIDEO_REFERENCE: 'video-to-video-reference';
+};
+export type KlingV3Operation = (typeof KlingV3Operation)[keyof typeof KlingV3Operation];
+export type KlingV3VideoGenInput = VideoGenInput & {
+  engine: 'kling-v3';
+} & {
+  operation?: KlingV3Operation;
+  mode?: KlingMode;
+  duration?: number;
+  aspectRatio?: KlingVideoGenAspectRatio;
+  /**
+   * Either A URL, A DataURL or a Base64 string
+   */
+  sourceImage?: string | null;
+  /**
+   * Either A URL, A DataURL or a Base64 string
+   */
+  endImage?: string | null;
+  videoUrl?: string | null;
+  images?: Array<string>;
+  elements?: Array<KlingV3ElementInput>;
+  generateAudio?: boolean;
+  voiceIds?: Array<string> | null;
+  keepAudio?: boolean;
+  multiPrompt?: Array<KlingV3MultiPrompt> | null;
+} & {
+  engine: 'kling-v3';
+};
 export declare const KlingVideoGenAspectRatio: {
   readonly '16:9': '16:9';
   readonly '9:16': '9:16';
@@ -1588,6 +2005,22 @@ export type MediaRatingInput = {
    * The engine to use for media rating. Valid values: "default" (HiveVLM) or "civitai".
    */
   engine?: string;
+  /**
+   * Include age classification analysis in the results (civitai engine only).
+   */
+  includeAgeClassification?: boolean;
+  /**
+   * Include face recognition and similarity analysis in the results (civitai engine only).
+   */
+  includeFaceRecognition?: boolean;
+  /**
+   * Include AI vs real image detection in the results (civitai engine only, GPU-only).
+   */
+  includeAIRecognition?: boolean;
+  /**
+   * Include anime vs real image detection in the results (civitai engine only, GPU-only).
+   */
+  includeAnimeRecognition?: boolean;
 };
 /**
  * Represents the output information returned from the MediaRating workflow step.
@@ -1606,6 +2039,10 @@ export type MediaRatingOutput = {
    * Detected content labels (e.g., "Animal", "Child", etc.).
    */
   labels?: Array<string> | null;
+  ageClassification?: AgeClassificationResult;
+  faceRecognition?: FaceRecognitionResult;
+  aiRecognition?: AiRecognitionResult;
+  animeRecognition?: AnimeRecognitionResult;
 };
 /**
  * MediaRating
@@ -2412,6 +2849,8 @@ export declare const SdCppSampleMethod: {
   readonly DDIM_TRAILING: 'ddim_trailing';
   readonly EULER_A: 'euler_a';
   readonly LCM: 'lcm';
+  readonly RES_MULTISTEP: 'res_multistep';
+  readonly RES_2S: 'res_2s';
 };
 export type SdCppSampleMethod = (typeof SdCppSampleMethod)[keyof typeof SdCppSampleMethod];
 export declare const SdCppSchedule: {
@@ -2420,6 +2859,7 @@ export declare const SdCppSchedule: {
   readonly KARRAS: 'karras';
   readonly EXPONENTIAL: 'exponential';
   readonly AYS: 'ays';
+  readonly BONG_TANGENT: 'bong_tangent';
 };
 export type SdCppSchedule = (typeof SdCppSchedule)[keyof typeof SdCppSchedule];
 /**
@@ -2500,6 +2940,23 @@ export type SoraVideoGenInput = VideoGenInput & {
 } & {
   engine: 'sora';
 };
+/**
+ * A system message that sets the behavior of the assistant.
+ */
+export type SystemMessage = ChatCompletionMessage & {
+  role: 'system';
+} & {
+  /**
+   * The system message content (text only).
+   */
+  content: string;
+  /**
+   * Optional name for the participant.
+   */
+  name?: string | null;
+} & {
+  role: 'system';
+};
 /**
  * Input for an text to image step.
  */
@@ -2607,6 +3064,52 @@ export type TextToImageStepTemplate = WorkflowStepTemplate & {
 } & {
   $type: 'textToImage';
 };
+export type TextToSpeechInput = {
+  engine: string | null;
+  /**
+   * The text to synthesize into speech.
+   */
+  text: string;
+  /**
+   * Target language (e.g., "English", "Chinese"). Defaults to "Auto".
+   */
+  language?: string | null;
+};
+/**
+ * Output from text-to-speech workflow step.
+ */
+export type TextToSpeechOutput = {
+  audioBlob: AudioBlob;
+  /**
+   * Whether the TTS used "base" (voice cloning) or "custom_voice" (built-in speaker) mode.
+   */
+  modelType?: string | null;
+  /**
+   * The speaker name used (set for CustomVoice mode, null for Base mode).
+   */
+  speaker?: string | null;
+};
+/**
+ * Text-to-Speech
+ */
+export type TextToSpeechStep = WorkflowStep & {
+  $type: 'textToSpeech';
+} & {
+  input: TextToSpeechInput;
+  output?: TextToSpeechOutput;
+} & {
+  $type: 'textToSpeech';
+};
+/**
+ * Text-to-Speech
+ */
+export type TextToSpeechStepTemplate = WorkflowStepTemplate & {
+  $type: 'textToSpeech';
+} & {
+  input: TextToSpeechInput;
+} & {
+  $type: 'textToSpeech';
+};
 /**
  * Represents training data in various formats
  */
@@ -2772,6 +3275,65 @@ export type TranscodeStepTemplate = WorkflowStepTemplate & {
 } & {
   $type: 'transcode';
 };
+export type TranscriptionInput = {
+  mediaUrl: string;
+  /**
+   * Optional language hint (e.g., "en", "zh") to guide transcription.
+   */
+  language?: string | null;
+  /**
+   * Optional context prompt to improve transcription accuracy.
+   */
+  context?: string | null;
+  /**
+   * Whether to return word-level timestamps. Defaults to true.
+   */
+  returnTimeStamps?: boolean;
+};
+export type TranscriptionOutput = {
+  /**
+   * The full transcribed text.
+   */
+  text: string;
+  /**
+   * The detected language of the audio.
+   */
+  language: string;
+  /**
+   * Word-level timestamps (if requested).
+   */
+  timeStamps?: Array<TranscriptionTimeStamp>;
+  /**
+   * Total processing time in seconds.
+   */
+  elapsedSeconds?: number;
+};
+/**
+ * Transcription
+ */
+export type TranscriptionStep = WorkflowStep & {
+  $type: 'transcription';
+} & {
+  input: TranscriptionInput;
+  output?: TranscriptionOutput;
+} & {
+  $type: 'transcription';
+};
+/**
+ * Transcription
+ */
+export type TranscriptionStepTemplate = WorkflowStepTemplate & {
+  $type: 'transcription';
+} & {
+  input: TranscriptionInput;
+} & {
+  $type: 'transcription';
+};
+export type TranscriptionTimeStamp = {
+  text: string;
+  startTime: number;
+  endTime: number;
+};
 export type TryOnUInput = {
   subjectUrl: string;
   garmentUrl: string;
@@ -2824,6 +3386,24 @@ export type UpdateWorkflowStepRequest = {
     [key: string]: unknown;
   };
 };
+/**
+ * A user message that can contain text and/or images.
+ */
+export type UserMessage = ChatCompletionMessage & {
+  role: 'user';
+} & {
+  /**
+   * The message content - can be a simple string or array of content parts.
+   * When deserialized from a string, it will be converted to a single TextContentPart.
+   */
+  content: Array<ChatCompletionContentPart>;
+  /**
+   * Optional name for the participant.
+   */
+  name?: string | null;
+} & {
+  role: 'user';
+};
 export type ValidationProblemDetails = {
   type?: string | null;
   title?: string | null;
@@ -2845,6 +3425,9 @@ export type ValidationProblemDetails = {
       }
     | undefined;
 };
+export type ValueTupleOfDoubleAndDouble = {
+  [key: string]: never;
+};
 export type ValueTupleOfStringAndInt32 = {
   [key: string]: never;
 };
@@ -2865,7 +3448,7 @@ export type Veo3VideoGenInput = VideoGenInput & {
   negativePrompt?: string | null;
   enablePromptEnhancer?: boolean;
   aspectRatio?: Veo3AspectRatio;
-  duration?: number;
+  duration?: 4 | 6 | 8;
   generateAudio?: boolean;
   seed?: number | null;
   fastMode?: boolean;
@@ -3937,6 +4520,9 @@ export type ZImageBaseImageGenInput = ZImageImageGenInput & {
   cfgScale?: number;
   seed?: number | null;
   quantity?: number;
+  loras?: {
+    [key: string]: number;
+  };
 } & {
   model: 'base';
 };
@@ -3967,6 +4553,9 @@ export type ZImageTurboImageGenInput = ZImageImageGenInput & {
   cfgScale?: number;
   seed?: number | null;
   quantity?: number;
+  loras?: {
+    [key: string]: number;
+  };
 } & {
   model: 'turbo';
 };
@@ -4173,6 +4762,35 @@ export type GetBlockedContentErrors = {
   403: ProblemDetails;
 };
 export type GetBlockedContentError = GetBlockedContentErrors[keyof GetBlockedContentErrors];
+export type InvokeAceStepAudioStepTemplateData = {
+  body?: AceStepAudioInput;
+  path?: never;
+  query?: {
+    experimental?: boolean;
+    allowMatureContent?: boolean;
+  };
+  url: '/v2/consumer/recipes/aceStepAudio';
+};
+export type InvokeAceStepAudioStepTemplateErrors = {
+  /**
+   * Bad Request
+   */
+  400: ProblemDetails;
+  /**
+   * Unauthorized
+   */
+  401: ProblemDetails;
+};
+export type InvokeAceStepAudioStepTemplateError =
+  InvokeAceStepAudioStepTemplateErrors[keyof InvokeAceStepAudioStepTemplateErrors];
+export type InvokeAceStepAudioStepTemplateResponses = {
+  /**
+   * OK
+   */
+  200: AceStepAudioOutput;
+};
+export type InvokeAceStepAudioStepTemplateResponse =
+  InvokeAceStepAudioStepTemplateResponses[keyof InvokeAceStepAudioStepTemplateResponses];
 export type InvokeAgeClassificationStepTemplateData = {
   body?: AgeClassificationInput;
   path?: never;
@@ -4550,6 +5168,35 @@ export type InvokeTextToImageStepTemplateResponses = {
 };
 export type InvokeTextToImageStepTemplateResponse =
   InvokeTextToImageStepTemplateResponses[keyof InvokeTextToImageStepTemplateResponses];
+export type InvokeTextToSpeechStepTemplateData = {
+  body?: TextToSpeechInput;
+  path?: never;
+  query?: {
+    experimental?: boolean;
+    allowMatureContent?: boolean;
+  };
+  url: '/v2/consumer/recipes/textToSpeech';
+};
+export type InvokeTextToSpeechStepTemplateErrors = {
+  /**
+   * Bad Request
+   */
+  400: ProblemDetails;
+  /**
+   * Unauthorized
+   */
+  401: ProblemDetails;
+};
+export type InvokeTextToSpeechStepTemplateError =
+  InvokeTextToSpeechStepTemplateErrors[keyof InvokeTextToSpeechStepTemplateErrors];
+export type InvokeTextToSpeechStepTemplateResponses = {
+  /**
+   * OK
+   */
+  200: TextToSpeechOutput;
+};
+export type InvokeTextToSpeechStepTemplateResponse =
+  InvokeTextToSpeechStepTemplateResponses[keyof InvokeTextToSpeechStepTemplateResponses];
 export type InvokeTrainingStepTemplateData = {
   body?: TrainingInput;
   path?: never;
@@ -4579,6 +5226,35 @@ export type InvokeTrainingStepTemplateResponses = {
 };
 export type InvokeTrainingStepTemplateResponse =
   InvokeTrainingStepTemplateResponses[keyof InvokeTrainingStepTemplateResponses];
+export type InvokeTranscriptionStepTemplateData = {
+  body?: TranscriptionInput;
+  path?: never;
+  query?: {
+    experimental?: boolean;
+    allowMatureContent?: boolean;
+  };
+  url: '/v2/consumer/recipes/transcription';
+};
+export type InvokeTranscriptionStepTemplateErrors = {
+  /**
+   * Bad Request
+   */
+  400: ProblemDetails;
+  /**
+   * Unauthorized
+   */
+  401: ProblemDetails;
+};
+export type InvokeTranscriptionStepTemplateError =
+  InvokeTranscriptionStepTemplateErrors[keyof InvokeTranscriptionStepTemplateErrors];
+export type InvokeTranscriptionStepTemplateResponses = {
+  /**
+   * OK
+   */
+  200: TranscriptionOutput;
+};
+export type InvokeTranscriptionStepTemplateResponse =
+  InvokeTranscriptionStepTemplateResponses[keyof InvokeTranscriptionStepTemplateResponses];
 export type InvokeVideoEnhancementStepTemplateData = {
   body?: VideoEnhancementInput;
   path?: never;