Multi-speaker Speech

Request parameters

FieldTypeRequiredDescription

model: stringRequired

Model ID. Use the stable model ID for regular calls, or a snapshot model ID when you need to reproduce results from a fixed version.

async: booleanOptional

When set to true, creates an asynchronous task and immediately returns a task object. Query the task for its result; webhook_url is only used for terminal-state notifications.

webhook_url: stringOptional

The HTTPS URL called back after the asynchronous task completes.

speakers: arrayRequired

The speaker configuration array, declaring each speaker and its voice; cannot be empty.

speakers[].id: stringRequired

The speaker identifier, referenced by segments[].speaker.

speakers[].voice_id: stringRequired

The speaker's voice ID; obtainable from the List Voices or Create Voice endpoint.

segments: arrayRequired

The segment array, giving the lines per speaker; cannot be empty.

segments[].speaker: stringRequired

The speaker identifier; must match one of speakers[].id.

segments[].text: stringRequired

The text of this line.

response_format: stringOptional

Audio format for synchronous requests. Asynchronous tasks may not honor this field; use the format returned by the task query as the source of truth.

delivery_method: stringOptional

Result delivery method.

aigc_metadata: objectOptional

Controls whether an invisible AIGC marker is written to the generated audio file. It is not written by default. This field affects only the metadata of the final audio file, not the synthesized speech content.

aigc_metadata.enabled: booleanOptional

Whether to write an invisible AIGC marker. Defaults to false.

aigc_metadata.content_propagator: stringOptional

The name or code of the content distribution service provider.

aigc_metadata.propagate_id: stringOptional

The content distribution ID used to identify this distribution.

Return value

When async=true is provided, the endpoint immediately returns an asynchronous task object. Use the returned task_id to query the result with GET /v1/audio/tasks/{task_id}.

Without asynchronous mode, delivery_method=url returns JSON containing an audio URL. When delivery_method=audio or the field is omitted, the endpoint returns audio binary data directly; the response header Content-Type determines its format.

Error codes

HTTP	error.code	Trigger	Suggested action
404	`voice_not_found`	A `speakers[].voice_id` does not exist or is unavailable	Verify each `speakers[].voice_id`.

Next steps

List Voices

List the voices available to your account.

Single-speaker Speech

Text-to-speech endpoint.

Request example

curl -X POST "https://api.mosi.cn/v1/audio/speech/speakers" \
  -H "Authorization: Bearer $MOSS_API_KEY" \
  -H "Platform: moss-platform" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "moss-ttsd-1.0",
  "speakers": [
    {
      "id": "speaker_a",
      "voice_id": "7e5020e3-b88e-4271-a996-2908a182dd32"
    },
    {
      "id": "speaker_b",
      "voice_id": "2925f9ce-75af-4354-8757-b52c3e541e8c"
    }
  ],
  "segments": [
    {
      "speaker": "speaker_a",
      "text": "诶，我最近看了一篇讲人工智能的文章，还挺有意思的，想跟你聊聊。"
    },
    {
      "speaker": "speaker_b",
      "text": "哦？是吗，关于啥的啊？又是哪个公司发了什么逆天的新模型吗？"
    }
  ],
  "response_format": "mp3",
  "delivery_method": "url"
}'

import os
import requests

response = requests.post(
        "https://api.mosi.cn/v1/audio/speech/speakers",
        headers={
        "Authorization": f"Bearer {os.environ['MOSS_API_KEY']}",
        "Platform": "moss-platform",
        "Content-Type": "application/json",
        },
        json={
            "model": "moss-ttsd-1.0",
            "speakers": [
                {
                    "id": "speaker_a",
                    "voice_id": "7e5020e3-b88e-4271-a996-2908a182dd32",
                },
                {
                    "id": "speaker_b",
                    "voice_id": "2925f9ce-75af-4354-8757-b52c3e541e8c",
                },
            ],
            "segments": [
                {
                    "speaker": "speaker_a",
                    "text": "诶，我最近看了一篇讲人工智能的文章，还挺有意思的，想跟你聊聊。",
                },
                {
                    "speaker": "speaker_b",
                    "text": "哦？是吗，关于啥的啊？又是哪个公司发了什么逆天的新模型吗？",
                },
            ],
            "response_format": "mp3",
            "delivery_method": "url",
        },
)

response.raise_for_status()
print(response.json())

const response = await fetch("https://api.mosi.cn/v1/audio/speech/speakers", {
  method: "POST",
  headers: {
    "Authorization": "Bearer " + process.env.MOSS_API_KEY,
    "Platform": "moss-platform",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "moss-ttsd-1.0",
    "speakers": [
      {
        "id": "speaker_a",
        "voice_id": "7e5020e3-b88e-4271-a996-2908a182dd32"
      },
      {
        "id": "speaker_b",
        "voice_id": "2925f9ce-75af-4354-8757-b52c3e541e8c"
      }
    ],
    "segments": [
      {
        "speaker": "speaker_a",
        "text": "诶，我最近看了一篇讲人工智能的文章，还挺有意思的，想跟你聊聊。"
      },
      {
        "speaker": "speaker_b",
        "text": "哦？是吗，关于啥的啊？又是哪个公司发了什么逆天的新模型吗？"
      }
    ],
    "response_format": "mp3",
    "delivery_method": "url"
  }),
})

if (!response.ok) throw new Error(`HTTP ${response.status}`)
console.log(await response.json())

Response example

{
  "id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "object": "audio.speech",
  "status": "SUCCESS",
  "url": "https://cdn.example.com/out.mp3",
  "response_format": "mp3",
  "content_type": "audio/mpeg",
  "created_at": 1710000000
}