Skip to content
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

add support for Kokoro via Replicate #1153

Merged
merged 9 commits into from
Feb 4, 2025
Merged

add support for Kokoro via Replicate #1153

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
01cdfe9
add support for Kokoro via Replicate
Vaibhavs10 Jan 29, 2025
3e04a18
add test
Vaibhavs10 Jan 29, 2025
30f4471
rename inputs to text for text-to-speech
hanouticelina Feb 3, 2025
40ef00b
remove outeTTS from Replicate supported models
hanouticelina Feb 3, 2025
cc280f6
rename only for replicate
hanouticelina Feb 3, 2025
033afe6
add OuteTTS back
hanouticelina Feb 3, 2025
341a42a
format
hanouticelina Feb 4, 2025
c8879b2
Merge branch 'main' into vb/replicate-kokoro
hanouticelina Feb 4, 2025
b45b0e4
Merge branch 'main' into vb/replicate-kokoro
hanouticelina Feb 4, 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
2 changes: 1 addition & 1 deletion packages/inference/src/providers/replicate.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ export const REPLICATE_SUPPORTED_MODEL_IDS: ProviderMapping<ReplicateId> = {
"stability-ai/sdxl:7762fd07cf82c948538e41f63f77d685e02b063e37e496e96eefd46c929f9bdc",
},
"text-to-speech": {
"OuteAI/OuteTTS-0.3-500M": "jbilcke/oute-tts:39a59319327b27327fa3095149c5a746e7f2aee18c75055c3368237a6503cd26",
"hexgrad/Kokoro-82M": "jaaari/kokoro-82m:dfdf537ba482b029e0a761699e6f55e9162cfd159270bfe0e44857caa5f275a6",
hanouticelina marked this conversation as resolved.
Show resolved Hide resolved
},
"text-to-video": {
"genmo/mochi-1-preview": "genmoai/mochi-1:1944af04d098ef69bed7f9d335d102e652203f268ec4aaa2d836f6217217e460",
Expand Down
20 changes: 15 additions & 5 deletions packages/inference/test/HfInference.spec.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,11 @@
import { expect, it, describe, assert } from "vitest";
import { assert, describe, expect, it } from "vitest";

import type { ChatCompletionStreamOutput } from "@huggingface/tasks";

import { chatCompletion, FAL_AI_SUPPORTED_MODEL_IDS, HfInference } from "../src";
import "./vcr";
import { readTestFile } from "./test-files";
import { textToVideo } from "../src/tasks/cv/textToVideo";
import { readTestFile } from "./test-files";
import "./vcr";

const TIMEOUT = 60000 * 3;
const env = import.meta.env;
Expand Down Expand Up @@ -939,11 +939,21 @@ describe.concurrent("HfInference", () => {
expect(res).toBeInstanceOf(Blob);
});

it("textToSpeech OuteTTS", async () => {
it.skip("textToSpeech OuteTTS - Needs to update the param name in jbilcke/oute-tts Replicate model to text instead of inputs", async () => {
const res = await client.textToSpeech({
hanouticelina marked this conversation as resolved.
Show resolved Hide resolved
hanouticelina marked this conversation as resolved.
Show resolved Hide resolved
model: "OuteAI/OuteTTS-0.3-500M",
provider: "replicate",
inputs: "OuteTTS is a frontier TTS model for its size of 1 Billion parameters",
text: "OuteTTS is a frontier TTS model for its size of 1 Billion parameters",
});

expect(res).toBeInstanceOf(Blob);
});

it("textToSpeech Kokoro", async () => {
const res = await client.textToSpeech({
model: "hexgrad/Kokoro-82M",
provider: "replicate",
text: "Kokoro is a frontier TTS model for its size of 1 Billion parameters",
});
Copy link
Contributor

Choose a reason for hiding this comment

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

Does it accept parameters?
Would be nice to add some to make sure they payload has the proper shape

Copy link
Contributor

Choose a reason for hiding this comment

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

yes, but it does not accept any of the parameters listed in TextToSpeechParameters (see the schema of the model here). Since each model on Replicate runs in its own docker container, models of the same task can have different parameters (and even more problematic: different input/output format). In huggingface_hub, we added an argument extra_body to allow users to pass any provider- or model-specific parameters.


expect(res).toBeInstanceOf(Blob);
Expand Down
8 changes: 4 additions & 4 deletions packages/tasks/src/tasks/text-to-speech/inference.ts
View file
Open in desktop
hanouticelina marked this conversation as resolved.
Show resolved Hide resolved
Original file line number Diff line number Diff line change
Expand Up @@ -21,14 +21,14 @@ export interface TextToSpeechOutput {
* Inputs for Text To Speech inference
*/
export interface TextToSpeechInput {
/**
* The input text data
*/
inputs: string;
/**
* Additional inference parameters for Text To Speech
*/
parameters?: TextToSpeechParameters;
/**
* The input text data
*/
text: string;
[property: string]: unknown;
}
/**
Expand Down
4 changes: 2 additions & 2 deletions packages/tasks/src/tasks/text-to-speech/spec/input.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
"title": "TextToSpeechInput",
"type": "object",
"properties": {
"inputs": {
"text": {
"description": "The input text data",
"type": "string"
},
Expand All @@ -26,5 +26,5 @@
}
}
},
"required": ["inputs"]
"required": ["text"]
}
Loading