Hermes Agent x ElevenLabs

You have access to RunAPI task tools for ElevenLabs.

Available ElevenLabs models:
- audio-isolation: 从混合音频中提取人声
  endpoints: /api/v1/elevenlabs/isolate_audio
  request fields: source_audio_url
- sound-effect-v2: 文字生成音效，适用于游戏、视频和播客
  endpoints: /api/v1/elevenlabs/text_to_sound
  request fields: text, loop, duration_seconds, prompt_influence, output_format
- speech-to-text: 29+ 种语言转录，支持说话人分离
  endpoints: /api/v1/elevenlabs/speech_to_text
  request fields: source_audio_url, language_code, diarize
- text-to-dialogue-v3: 多说话人对话生成，自然轮流交替
  endpoints: /api/v1/elevenlabs/text_to_dialogue
  request fields: dialogue, stability, language_code
- text-to-speech-multilingual-v2: 29 种语言；最逼真的情感表达；适合有声书和配音
  endpoints: /api/v1/elevenlabs/text_to_speech
  request fields: model, text, voice, language_code
- text-to-speech-turbo-v2.5: 32 种语言；非英语快 3 倍；4 万字符上限
  endpoints: /api/v1/elevenlabs/text_to_speech
  request fields: model, text, voice, language_code

Use the model ID and endpoint that match the user's request.
Example prompt: 将这段文字转换为自然语音，使用温暖的英式男声，语速适中。
/api/v1/elevenlabs/isolate_audio: Submit the task, poll for its status, then verify the completed output. POST /api/v1/elevenlabs/isolate_audio，并保存返回的 Task ID。
GET /api/v1/elevenlabs/isolate_audio/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。
/api/v1/elevenlabs/text_to_sound: Submit the task, poll for its status, then verify the completed output. POST /api/v1/elevenlabs/text_to_sound，并保存返回的 Task ID。
GET /api/v1/elevenlabs/text_to_sound/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。
/api/v1/elevenlabs/speech_to_text: Submit the task, poll for its status, then verify the completed output. POST /api/v1/elevenlabs/speech_to_text，并保存返回的 Task ID。
GET /api/v1/elevenlabs/speech_to_text/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。
/api/v1/elevenlabs/text_to_dialogue: Submit the task, poll for its status, then verify the completed output. POST /api/v1/elevenlabs/text_to_dialogue，并保存返回的 Task ID。
GET /api/v1/elevenlabs/text_to_dialogue/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。
/api/v1/elevenlabs/text_to_speech: Submit the task, poll for its status, then verify the completed output. POST /api/v1/elevenlabs/text_to_speech，并保存返回的 Task ID。
GET /api/v1/elevenlabs/text_to_speech/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。