Get Task · Moss API

Path parameters

FieldTypeRequiredDescription

task_idstringRequired

The task ID returned by the asynchronous request.

Return value

The endpoint returns the task's current status. On success, result fields appear directly at the top level of the response.

FieldTypeDescription

id: string

Task ID corresponding to the task_id in the path.

object: string

Task object type, for example audio.speech / audio.voice_generation / audio.transcription.

status: string

Task status, for example PENDING / PROCESSING / SUCCESS / FAILED.

created_at: integer

Creation time, a Unix timestamp in seconds.

updated_at: integer

Last update time, a Unix timestamp in seconds.

error: object

Task execution failure details.

Audio task result

When status=SUCCESS and object is audio.speech or audio.voice_generation, the response includes an audio result URL. Use the returned response_format and content_type for the actual format.

FieldTypeConditionDescription

urlstringOn success

Audio result URL.

response_formatstringOn success

Actual output audio format, for example wav.

content_typestringOn success

Audio MIME type, for example audio/wav.

Transcription task result

When status=SUCCESS and object=audio.transcription, text contains the complete transcription; speaker diarization tasks also return segments.

FieldTypeConditionDescription

textstringSuccessful standard transcription

Complete transcription text.

segmentsarraySuccessful diarization

Diarized segments; may be an empty array when there are no valid segments.

segments[].speakerstringWith segments

Speaker identifier.

segments[].textstringWith segments

Text for the segment.

segments[].startnumberWith segments

Segment start time in seconds.

segments[].endnumberWith segments

Segment end time in seconds.

Task failure

status=FAILED means the task was created but failed during execution. This differs from a query error when the task does not exist or cannot be accessed. The failure details are in error.

Error codes

HTTP	error.code	Trigger	Suggested action
404	`task_not_found`	The specified `task_id` does not exist or is not visible	Use the `task_id` returned by the creation endpoint and verify that the API Key can access the task.

Next steps

Single-speaker Speech

Text-to-speech endpoint.

Create Transcription

Audio transcription endpoint.

Request example

curl -X GET "https://api.mosi.cn/v1/audio/tasks/<task_id>" \
  -H "Authorization: Bearer $MOSS_API_KEY"

import os
import requests

response = requests.get(
        "https://api.mosi.cn/v1/audio/tasks/<task_id>",
        headers={
        "Authorization": f"Bearer {os.environ['MOSS_API_KEY']}",
        },
)

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

const response = await fetch("https://api.mosi.cn/v1/audio/tasks/<task_id>", {
  method: "GET",
  headers: {
    "Authorization": "Bearer " + process.env.MOSS_API_KEY,
  },
})

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",
  "created_at": 1710000000,
  "updated_at": 1710000003,
  "url": "https://example.com/result.wav",
  "response_format": "wav",
  "content_type": "audio/wav"
}

{
  "id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "object": "audio.transcription",
  "status": "SUCCESS",
  "created_at": 1710000000,
  "updated_at": 1710000002,
  "text": "This is a short transcription test audio."
}

{
  "id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "object": "audio.voice_generation",
  "status": "PROCESSING",
  "created_at": 1710000000,
  "updated_at": 1710000001
}

{
  "id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "object": "audio.speech",
  "status": "FAILED",
  "error": {},
  "created_at": 1710000000,
  "updated_at": 1710000001
}