Skip to content

[Inference] Add ASR support for Replicate provider #1679

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Aug 25, 2025
Merged

[Inference] Add ASR support for Replicate provider #1679

Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
ceaf141
Add ASR support for Replicate provider
lucataco Aug 7, 2025
af7a21e
Fix lint
lucataco Aug 7, 2025
ba1755a
Merge branch 'main' into main
lucataco Aug 11, 2025
5299c1c
Add ReplicateASR mapping
lucataco Aug 12, 2025
de2778d
fix get response
hanouticelina Aug 12, 2025
3dae24c
Merge branch 'main' into main
coyotte508 Aug 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 58 additions & 0 deletions packages/inference/src/providers/replicate.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,11 +20,14 @@ import type { BodyParams, HeaderParams, RequestArgs, UrlParams } from "../types.
import { omit } from "../utils/omit.js";
import {
TaskProviderHelper,
type AutomaticSpeechRecognitionTaskHelper,
type ImageToImageTaskHelper,
type TextToImageTaskHelper,
type TextToVideoTaskHelper,
} from "./providerHelper.js";
import type { ImageToImageArgs } from "../tasks/cv/imageToImage.js";
import type { AutomaticSpeechRecognitionArgs } from "../tasks/audio/automaticSpeechRecognition.js";
import type { AutomaticSpeechRecognitionOutput } from "@huggingface/tasks";
import { base64FromBytes } from "../utils/base64FromBytes.js";
export interface ReplicateOutput {
output?: string | string[];
Expand Down Expand Up @@ -163,6 +166,61 @@ export class ReplicateTextToVideoTask extends ReplicateTask implements TextToVid
}
}

export class ReplicateAutomaticSpeechRecognitionTask
extends ReplicateTask
implements AutomaticSpeechRecognitionTaskHelper
{
override preparePayload(params: BodyParams): Record<string, unknown> {
return {
input: {
...omit(params.args, ["inputs", "parameters"]),
...(params.args.parameters as Record<string, unknown>),
audio: params.args.inputs, // This will be processed in preparePayloadAsync
},
version: params.model.includes(":") ? params.model.split(":")[1] : undefined,
};
}

async preparePayloadAsync(args: AutomaticSpeechRecognitionArgs): Promise<RequestArgs> {
const blob = "data" in args && args.data instanceof Blob ? args.data : "inputs" in args ? args.inputs : undefined;

if (!blob || !(blob instanceof Blob)) {
throw new Error("Audio input must be a Blob");
Comment on lines +187 to +188
Copy link
Preview

Copilot AI Aug 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The error message 'Audio input must be a Blob' is not descriptive enough. Consider providing more context about expected input formats and how to convert them to Blob.

Suggested change
if (!blob || !(blob instanceof Blob)) {
throw new Error("Audio input must be a Blob");
throw new Error(
"Audio input must be a Blob (e.g., a File or Blob object from the browser). " +
"Received: " + (blob === undefined ? "undefined" : typeof blob) + ". " +
"To convert an ArrayBuffer or base64 string to a Blob, use: " +
"`new Blob([arrayBuffer], { type: 'audio/wav' })` or " +
"`fetch('data:audio/wav;base64,...').then(res => res.blob())`. " +
"See documentation for supported input formats."
);

Copilot uses AI. Check for mistakes.

}

// Convert Blob to base64 data URL
const bytes = new Uint8Array(await blob.arrayBuffer());
const base64 = base64FromBytes(bytes);
const audioInput = `data:${blob.type || "audio/wav"};base64,${base64}`;

return {
...("data" in args ? omit(args, "data") : omit(args, "inputs")),
inputs: audioInput,
};
}

override async getResponse(response: ReplicateOutput): Promise<AutomaticSpeechRecognitionOutput> {
if (typeof response === "object" && !!response && "output" in response && typeof response.output === "string") {
return { text: response.output };
}

if (
typeof response === "object" &&
!!response &&
"output" in response &&
Array.isArray(response.output) &&
response.output.length > 0 &&
typeof response.output[0] === "string"
) {
return { text: response.output[0] };
}

throw new InferenceClientProviderOutputError(
"Received malformed response from Replicate automatic-speech-recognition API"
);
}
}

export class ReplicateImageToImageTask extends ReplicateTask implements ImageToImageTaskHelper {
override preparePayload(params: BodyParams<ImageToImageArgs>): Record<string, unknown> {
return {
Expand Down