@huggingface/tasks 0.13.0-test2 → 0.13.0-test4

package/dist/tasks/text-to-audio/inference.d.cts

@@ -0,0 +1,139 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Audio inference
+ */
+export interface TextToAudioInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToAudioParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Audio
+ */
+export interface TextToAudioParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generation_parameters?: GenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+export interface GenerationParameters {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over max_length.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over min_length.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+export type EarlyStoppingUnion = boolean | "never";
+/**
+ * Outputs of inference for the Text To Audio task
+ */
+export interface TextToAudioOutput {
+    /**
+     * The generated audio waveform.
+     */
+    audio: unknown;
+    samplingRate: unknown;
+    /**
+     * The sampling rate of the generated audio waveform.
+     */
+    sampling_rate?: number;
+    [property: string]: unknown;
+}
+//# sourceMappingURL=inference.d.cts.map

package/dist/tasks/text-to-audio/inference.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"inference.d.cts","sourceRoot":"","sources":["../../../src/tasks/text-to-audio/inference.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAChC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;OAEG;IACH,UAAU,CAAC,EAAE,qBAAqB,CAAC;IACnC,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,qBAAqB;IACrC;;OAEG;IACH,qBAAqB,CAAC,EAAE,oBAAoB,CAAC;IAC7C,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,oBAAoB;IACpC;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,kBAAkB,CAAC;IACpC;;;;;OAKG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;;;;;;OAQG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;;OAGG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB;;OAEG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;OAEG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;;;;OAMG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,MAAM,kBAAkB,GAAG,OAAO,GAAG,OAAO,CAAC;AAEnD;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,KAAK,EAAE,OAAO,CAAC;IACf,YAAY,EAAE,OAAO,CAAC;IACtB;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B"}

package/dist/tasks/text-to-audio/inference.d.ts

@@ -0,0 +1,139 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Audio inference
+ */
+export interface TextToAudioInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToAudioParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Audio
+ */
+export interface TextToAudioParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generation_parameters?: GenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+export interface GenerationParameters {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over max_length.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over min_length.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+export type EarlyStoppingUnion = boolean | "never";
+/**
+ * Outputs of inference for the Text To Audio task
+ */
+export interface TextToAudioOutput {
+    /**
+     * The generated audio waveform.
+     */
+    audio: unknown;
+    samplingRate: unknown;
+    /**
+     * The sampling rate of the generated audio waveform.
+     */
+    sampling_rate?: number;
+    [property: string]: unknown;
+}
+//# sourceMappingURL=inference.d.ts.map

package/dist/tasks/text-to-audio/inference.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"inference.d.ts","sourceRoot":"","sources":["../../../src/tasks/text-to-audio/inference.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAChC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;OAEG;IACH,UAAU,CAAC,EAAE,qBAAqB,CAAC;IACnC,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,qBAAqB;IACrC;;OAEG;IACH,qBAAqB,CAAC,EAAE,oBAAoB,CAAC;IAC7C,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,oBAAoB;IACpC;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,kBAAkB,CAAC;IACpC;;;;;OAKG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;;;;;;OAQG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;;OAGG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB;;OAEG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;OAEG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;;;;OAMG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,MAAM,kBAAkB,GAAG,OAAO,GAAG,OAAO,CAAC;AAEnD;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,KAAK,EAAE,OAAO,CAAC;IACf,YAAY,EAAE,OAAO,CAAC;IACtB;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B"}

package/dist/tasks/text-to-image/data.d.cts

@@ -0,0 +1,4 @@
+import type { TaskDataCustom } from "..";
+declare const taskData: TaskDataCustom;
+export default taskData;
+//# sourceMappingURL=data.d.cts.map

package/dist/tasks/text-to-image/data.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"data.d.cts","sourceRoot":"","sources":["../../../src/tasks/text-to-image/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,IAAI,CAAC;AAEzC,QAAA,MAAM,QAAQ,EAAE,cA+Ff,CAAC;AAEF,eAAe,QAAQ,CAAC"}

package/dist/tasks/text-to-image/data.d.ts

@@ -0,0 +1,4 @@
+import type { TaskDataCustom } from "..";
+declare const taskData: TaskDataCustom;
+export default taskData;
+//# sourceMappingURL=data.d.ts.map

package/dist/tasks/text-to-image/data.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"data.d.ts","sourceRoot":"","sources":["../../../src/tasks/text-to-image/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,IAAI,CAAC;AAEzC,QAAA,MAAM,QAAQ,EAAE,cA+Ff,CAAC;AAEF,eAAe,QAAQ,CAAC"}

package/dist/tasks/text-to-image/inference.d.cts

@@ -0,0 +1,72 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Image inference
+ */
+export interface TextToImageInput {
+    /**
+     * The input text data (sometimes called "prompt")
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToImageParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Image
+ */
+export interface TextToImageParameters {
+    /**
+     * A higher guidance scale value encourages the model to generate images closely linked to
+     * the text prompt, but values too high may cause saturation and other artifacts.
+     */
+    guidance_scale?: number;
+    /**
+     * One or several prompt to guide what NOT to include in image generation.
+     */
+    negative_prompt?: string[];
+    /**
+     * The number of denoising steps. More denoising steps usually lead to a higher quality
+     * image at the expense of slower inference.
+     */
+    num_inference_steps?: number;
+    /**
+     * Override the scheduler with a compatible one.
+     */
+    scheduler?: string;
+    /**
+     * Seed for the random number generator.
+     */
+    seed?: number;
+    /**
+     * The size in pixel of the output image
+     */
+    target_size?: TargetSize;
+    [property: string]: unknown;
+}
+/**
+ * The size in pixel of the output image
+ */
+export interface TargetSize {
+    height: number;
+    width: number;
+    [property: string]: unknown;
+}
+/**
+ * Outputs of inference for the Text To Image task
+ */
+export interface TextToImageOutput {
+    /**
+     * The generated image returned as raw bytes in the payload.
+     */
+    image: unknown;
+    [property: string]: unknown;
+}
+//# sourceMappingURL=inference.d.cts.map

package/dist/tasks/text-to-image/inference.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"inference.d.cts","sourceRoot":"","sources":["../../../src/tasks/text-to-image/inference.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAChC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;OAEG;IACH,UAAU,CAAC,EAAE,qBAAqB,CAAC;IACnC,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,qBAAqB;IACrC;;;OAGG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;OAEG;IACH,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;IAC3B;;;OAGG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAC;IAC7B;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;OAEG;IACH,IAAI,CAAC,EAAE,MAAM,CAAC;IACd;;OAEG;IACH,WAAW,CAAC,EAAE,UAAU,CAAC;IACzB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,WAAW,UAAU;IAC1B,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;IACd,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,KAAK,EAAE,OAAO,CAAC;IACf,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B"}

package/dist/tasks/text-to-image/inference.d.ts

@@ -0,0 +1,72 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Image inference
+ */
+export interface TextToImageInput {
+    /**
+     * The input text data (sometimes called "prompt")
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToImageParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Image
+ */
+export interface TextToImageParameters {
+    /**
+     * A higher guidance scale value encourages the model to generate images closely linked to
+     * the text prompt, but values too high may cause saturation and other artifacts.
+     */
+    guidance_scale?: number;
+    /**
+     * One or several prompt to guide what NOT to include in image generation.
+     */
+    negative_prompt?: string[];
+    /**
+     * The number of denoising steps. More denoising steps usually lead to a higher quality
+     * image at the expense of slower inference.
+     */
+    num_inference_steps?: number;
+    /**
+     * Override the scheduler with a compatible one.
+     */
+    scheduler?: string;
+    /**
+     * Seed for the random number generator.
+     */
+    seed?: number;
+    /**
+     * The size in pixel of the output image
+     */
+    target_size?: TargetSize;
+    [property: string]: unknown;
+}
+/**
+ * The size in pixel of the output image
+ */
+export interface TargetSize {
+    height: number;
+    width: number;
+    [property: string]: unknown;
+}
+/**
+ * Outputs of inference for the Text To Image task
+ */
+export interface TextToImageOutput {
+    /**
+     * The generated image returned as raw bytes in the payload.
+     */
+    image: unknown;
+    [property: string]: unknown;
+}
+//# sourceMappingURL=inference.d.ts.map

package/dist/tasks/text-to-image/inference.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"inference.d.ts","sourceRoot":"","sources":["../../../src/tasks/text-to-image/inference.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAChC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;OAEG;IACH,UAAU,CAAC,EAAE,qBAAqB,CAAC;IACnC,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,qBAAqB;IACrC;;;OAGG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;OAEG;IACH,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;IAC3B;;;OAGG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAC;IAC7B;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;OAEG;IACH,IAAI,CAAC,EAAE,MAAM,CAAC;IACd;;OAEG;IACH,WAAW,CAAC,EAAE,UAAU,CAAC;IACzB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,WAAW,UAAU;IAC1B,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;IACd,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,KAAK,EAAE,OAAO,CAAC;IACf,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B"}

package/dist/tasks/text-to-speech/data.d.cts

@@ -0,0 +1,4 @@
+import type { TaskDataCustom } from "..";
+declare const taskData: TaskDataCustom;
+export default taskData;
+//# sourceMappingURL=data.d.cts.map

package/dist/tasks/text-to-speech/data.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"data.d.cts","sourceRoot":"","sources":["../../../src/tasks/text-to-speech/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,IAAI,CAAC;AAEzC,QAAA,MAAM,QAAQ,EAAE,cA0Ef,CAAC;AAEF,eAAe,QAAQ,CAAC"}

package/dist/tasks/text-to-speech/data.d.ts

@@ -0,0 +1,4 @@
+import type { TaskDataCustom } from "..";
+declare const taskData: TaskDataCustom;
+export default taskData;
+//# sourceMappingURL=data.d.ts.map

package/dist/tasks/text-to-speech/data.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"data.d.ts","sourceRoot":"","sources":["../../../src/tasks/text-to-speech/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,IAAI,CAAC;AAEzC,QAAA,MAAM,QAAQ,EAAE,cA0Ef,CAAC;AAEF,eAAe,QAAQ,CAAC"}

package/dist/tasks/text-to-speech/inference.d.cts

@@ -0,0 +1,141 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Speech inference
+ */
+export interface TextToSpeechInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToSpeechParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Speech
+ */
+export interface TextToSpeechParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generation_parameters?: GenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+export interface GenerationParameters {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over max_length.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over min_length.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+export type EarlyStoppingUnion = boolean | "never";
+/**
+ * Outputs for Text to Speech inference
+ *
+ * Outputs of inference for the Text To Audio task
+ */
+export interface TextToSpeechOutput {
+    /**
+     * The generated audio waveform.
+     */
+    audio: unknown;
+    samplingRate: unknown;
+    /**
+     * The sampling rate of the generated audio waveform.
+     */
+    sampling_rate?: number;
+    [property: string]: unknown;
+}
+//# sourceMappingURL=inference.d.cts.map

package/dist/tasks/text-to-speech/inference.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"inference.d.cts","sourceRoot":"","sources":["../../../src/tasks/text-to-speech/inference.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAEH;;GAEG;AACH,MAAM,WAAW,iBAAiB;IACjC;;OAEG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;OAEG;IACH,UAAU,CAAC,EAAE,sBAAsB,CAAC;IACpC,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,sBAAsB;IACtC;;OAEG;IACH,qBAAqB,CAAC,EAAE,oBAAoB,CAAC;IAC7C,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;;;GAIG;AACH,MAAM,WAAW,oBAAoB;IACpC;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,kBAAkB,CAAC;IACpC;;;;;OAKG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;;;;;;OAQG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;OAEG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;;OAGG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB;;OAEG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;OAEG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,KAAK,CAAC,EAAE,MAAM,CAAC;IACf;;;;;;OAMG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B;AAED;;GAEG;AACH,MAAM,MAAM,kBAAkB,GAAG,OAAO,GAAG,OAAO,CAAC;AAEnD;;;;GAIG;AACH,MAAM,WAAW,kBAAkB;IAClC;;OAEG;IACH,KAAK,EAAE,OAAO,CAAC;IACf,YAAY,EAAE,OAAO,CAAC;IACtB;;OAEG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC;CAC5B"}