Use Seed Audio 1.0 Multilingual with Comfy Router

Call byteplus/seed-audio-1.0-multilingual through Comfy Router: endpoint, request shape and the response Router returns.

API Reference for byteplus/seed-audio-1.0-multilingual, served by Comfy Router from BytePlus.

Quick start

Create a key at platform.comfy.org/profile/api-keys and export it as COMFY_API_KEY. The Python and TypeScript snippets use the Comfy SDKs (pip install comfy-sdk, npm install @comfyorg/sdk); the cURL snippet is the same call over raw HTTP.

Model ID: byteplus/seed-audio-1.0-multilingual

Endpoint: POST https://api.comfy.org/v2/models/byteplus/seed-audio-1.0-multilingual

Python

from comfy_sdk import Comfy

# Reads COMFY_API_KEY from the environment. Each call sends a fresh
# Idempotency-Key and waits up to 10 minutes for the finished result.
with Comfy() as client:
    result = client.models.run(
        "byteplus/seed-audio-1.0-multilingual",
        {
            "audio_config": {
                "format": "wav",
                "sample_rate": 24000,
            },
            "text_prompt": "Hello from Comfy Router.",
        },
    )

print(result)

Schema

Input

audio_config · object

Output audio configuration.

audio_config.enable_subtitle · boolean

Whether to enable the subtitle service. When enabled, the response includes utterance- and word-level timestamps (default: false).

audio_config.format · string

Output audio format: wav (default), mp3, pcm or ogg_opus.

Possible values: wav, mp3, pcm, ogg_opus

audio_config.loudness_rate · integer

-50 to 100; 100 means 2.0x volume, -50 means 0.5x volume (default: 0).

Range: -50 to 100

audio_config.pitch_rate · integer

-12 to 12 (default: 0).

Range: -12 to 12

audio_config.sample_rate · integer

Output sample rate: 8000, 16000, 24000 (default), 32000, 44100 or 48000.

Possible values: 8000, 16000, 24000, 32000, 44100, 48000

audio_config.speech_rate · integer

-50 to 100; 100 means 2.0x speed, -50 means 0.5x speed (default: 0).

Range: -50 to 100

model · string

Model identifier. Supported models: seed-audio-1.0 and seed-audio-1.0-multilingual. A direct v1 call to POST /proxy/byteplus/api/v3/tts/create MUST supply it — the proxy refuses any other value, and an omitted one, with a 400. It is NOT in this schema’s required list because Comfy Router fills it from the {model} path segment of /v2/models/byteplus/{model}, so a Router caller omits it.

references · object[]

Reference resources. Omit for text-only generation. Up to 3 audio references (each up to 30 seconds and 10 MB decoded; wav, mp3, pcm or ogg_opus) or exactly 1 image reference (up to 10 MB decoded; jpeg, png or webp). Image references cannot be mixed with audio references. Those ceilings are BytePlus’s and are stated in DECODED bytes: base64 inflates an asset by about 4/3, and a Comfy Router call is capped at 10 MiB for the whole JSON document, so an inline audio_data/image_data reference sent to /v2/models/byteplus/{model} has to stay under roughly 7.5 MB decoded. Use audio_url/image_url for anything larger — a reference at BytePlus’s own 10 MB maximum is refused 413 before validation runs if it is inlined.

references[].audio_data · string

Base64-encoded reference audio.

references[].audio_url · string

URL of a remote reference audio file.

references[].image_data · string

Base64-encoded reference image.

references[].image_url · string

URL of a remote reference image.

references[].speaker · string

Voice ID. Can be a supported Doubao TTS voice or a voice-clone voice ID.

text_prompt · string · required

Prompt or text to synthesize (1 to 3,000 characters; an empty string is refused). When audio references are provided, reference them by order using @Audio1, @Audio2 and @Audio3.

watermark · object

Watermark configuration object. An empty object is accepted.

watermark.aigc_metadata · object

Implicit watermark. Adds metadata to the synthesized audio header.

watermark.aigc_metadata.content_producer · string

Name or code of the synthesis service provider.

watermark.aigc_metadata.content_propagator · string

Name or code of the content distribution service provider.

watermark.aigc_metadata.enable · boolean

Whether to enable the implicit watermark (default: false).

watermark.aigc_metadata.produce_id · string

Content production ID.

watermark.aigc_metadata.propagate_id · string

Content distribution ID.

watermark.aigc_watermark · boolean

Explicit watermark switch. Adds an audio rhythm marker at the end of the synthesized audio (default: false).

Generated from the schema Router serves at GET /v2/models/byteplus/seed-audio-1.0-multilingual/openapi.json, the same document it validates a call against before the request reaches the provider.

Output

audio · string

Generated audio data, Base64-encoded.

code · integer

Status code. Refer to the official error-code document for details.

duration · number

Duration after speed or post-processing, in seconds.

Format: double

message · string

Status details.

original_duration · number

Original model output duration in seconds. Used for billing and capped at 120 seconds.

Format: double

subtitle · object

Subtitle information for the audio. Present only when audio_config.enable_subtitle is set to true in the request.

subtitle.sentences · object[]

Utterance-level subtitle information.

subtitle.sentences[].end_time · integer

Segment end time, in milliseconds from the beginning of the audio.

subtitle.sentences[].start_time · integer

Segment start time, in milliseconds from the beginning of the audio.

subtitle.sentences[].text · string

The complete text of the segment.

subtitle.text · string

Subtitle text corresponding to the audio.

subtitle.words · object[]

Word-level subtitle information.

subtitle.words[].end_time · integer

Segment end time, in milliseconds from the beginning of the audio.

subtitle.words[].start_time · integer

Segment start time, in milliseconds from the beginning of the audio.

subtitle.words[].text · string

The complete text of the segment.

url · string

Temporary audio URL, valid for 2 hours.

Examples

Input

{
  "audio_config": {
    "format": "wav",
    "sample_rate": 24000
  },
  "text_prompt": "Hello from Comfy Router."
}

Output

{
  "audio": "PGJhc2U2ND4=",
  "duration": 1.2,
  "original_duration": 1.2,
  "url": "https://example.invalid/byteplus/seed-audio/generated.wav"
}

Before you ship

The snippets above are the shortest working call. Three things are the same for every model and are documented once on the Comfy Router headers page: send an Idempotency-Key on every paid call and reuse it when you retry, expect the connection to be held up to Router’s 10 minute deadline, and keep X-Comfy-Request-Id from every response. The SDKs do all three for you; the cURL tab does none of them. On failure, X-Comfy-Error-Type names the bucket, and a 422 means the body failed the model’s schema and was never billed.

Headers

Authentication, idempotency, request IDs, error buckets, retry pacing, spend limits.

Quick Start

Typed error handling in Python and TypeScript, reading the 422, walking the catalog.

Limitations

What Router does not do today, and what to use instead.