Multi-Speaker Dialogue Speech Generation

Synthesize multi-character dialogue text into complete dialogue audio, suited to podcasts, role-play, audiobooks, and course dialogues.

Workflow

1. Create an API Key

   Open the API Key management platform, create an API Key, and save it as an environment variable. Do not write the API Key into client-side code or commit it to a repository.

   export MOSS_API_KEY="<your-api-key>"
   

2. Prepare the request input

   Use model to specify the model, use speakers to declare each speaker (id + a created voice_id), then use segments to provide the lines split by speaker.

   speakers: mapping of speaker id to voice_id segments: lines split by speaker delivery_method and response_format: control whether a URL or audio is returned

   {
     "model": "moss-ttsd",
     "speakers": [
       { "id": "A", "voice_id": "<speaker_a_voice_id>" },
       { "id": "B", "voice_id": "<speaker_b_voice_id>" }
     ],
     "segments": [
       { "speaker": "A", "text": "How is the progress today?" },
       { "speaker": "B", "text": "The API integration testing is already complete." }
     ],
     "delivery_method": "url",
     "response_format": "mp3"
   }
   

3. Send the first request

   Send a request to POST /v1/audio/speech/speakers. When delivery_method=url, JSON and a result URL are returned synchronously.

   cURLPythonTypeScript

   curl https://api.mosi.cn/v1/audio/speech/speakers \
     -H "Authorization: Bearer $MOSS_API_KEY" \
     -H "Content-Type: application/json" \
     -d @request.json
   

   import json
   import os
   import requests
   
   payload = json.load(open("request.json"))
   resp = requests.post(
       "https://api.mosi.cn/v1/audio/speech/speakers",
       headers={"Authorization": f"Bearer {os.environ['MOSS_API_KEY']}"},
       json=payload,
   )
   print(resp.json())
   

   const resp = await fetch('https://api.mosi.cn/v1/audio/speech/speakers', {
     method: 'POST',
     headers: {
       Authorization: `Bearer ${apiKey}`,
       'Content-Type': 'application/json',
     },
     body: JSON.stringify(payload),
   })
   

4. Retrieve the result

   When delivery_method=url, read the result URL from the JSON of the synchronous response. For longer tasks, you can pass async=true and webhook_url to create an asynchronous task, then query it with the returned task_id.

   curl https://api.mosi.cn/v1/audio/tasks/$TASK_ID \
     -H "Authorization: Bearer $MOSS_API_KEY"
   

5. Verify success

   Success indicator: you obtain the complete dialogue audio (delivery_method=audio) or a result URL (delivery_method=url), and can download or play it.

Next steps