Give an approved script a voice
Turn approved text into narration and spoken audio.
MCP / oo CLI + Skill4 個接入服務
可以做什麼
Prepare a script and choose a supported voice and format. Speech generation helps create narration for a video, an explanation, or a listening version of written material.
Create a narration track
Pair a reviewed script with a supported voice and retrieve the resulting audio.
Read a short explanation
Turn a concise piece of approved text into spoken audio.
Audition an original voice
Generate short voice candidates from a description and preview text.
Generate speech
Turn text into speech with Qwen.
試試這些提示詞
替換方括號裡的內容,複製到已接入 OOMOL 的 AI 助手。
Create a narration track
Pair a reviewed script with a supported voice and retrieve the resulting audio.
Use OOMOL MCP with Doubao Speech to complete this task: Use Doubao Speech to narrate [approved script] with [voice and format]. Confirm the script and voice, submit the task, and retrieve the audio.
Read a short explanation
Turn a concise piece of approved text into spoken audio.
Use OOMOL MCP with Qwen to complete this task: Use Qwen to turn [approved explanation] into speech using a supported voice. Confirm the voice and output format before generating.
Audition an original voice
Generate short voice candidates from a description and preview text.
Use OOMOL MCP with AIsa to complete this task: Use AIsa to generate two original voice previews described as [voice direction], speaking [short preview text]. Return the WAV audio candidates.
Generate speech
Turn text into speech with Qwen.
Use OOMOL MCP with OOMOL Fusion API to complete this task: Use OOMOL Qwen TTS to read [text] with [voice].
開始使用 Text to Speech
接入你的 AI 助手
在 MCP 用戶端中新增 OOMOL,使用你的團隊完成登入,即可呼叫接入服務。
安裝 oo CLI 與 Skill,完成 OOMOL 登入後,在 AI 助手中使用 /oo。
安裝 oo CLI選擇服務
Doubao SpeechChoose OOMOL-provided access when available, or connect your Doubao Speech account.
Pair a reviewed script with a supported voice and retrieve the resulting audio.
查看 Doubao Speech 接入說明 →QwenChoose OOMOL-provided access when available, or connect your Qwen account.
Turn a concise piece of approved text into spoken audio.
查看 Qwen 接入說明 →AIsaChoose OOMOL-provided access when available, or connect your AIsa account.
Generate short Fish Audio voice previews without creating a saved voice model.
查看 AIsa 接入說明 →OOMOL Fusion APISign in to OOMOL; no separate provider credential is required.
Use OOMOL services with your current team.
查看 OOMOL Fusion API 接入說明 →嘗試第一個任務
Use OOMOL MCP with Doubao Speech to complete this task: Read the tool definitions provided by Doubao Speech for this Text to Speech task. List the required inputs and supported settings so I can prepare a small example. Inspect definitions only; do not execute the task.
提供 Text to Speech 能力的服務
依任務選擇服務。操作數為各服務的全部操作總數。
Doubao Speech
Pair a reviewed script with a supported voice and retrieve the resulting audio.
Qwen
Turn a concise piece of approved text into spoken audio.
AIsa
Generate short Fish Audio voice previews without creating a saved voice model.
OOMOL Fusion API
Use OOMOL services with your current team.
所有操作
Submit Tts
Submit an asynchronous Doubao Speech 2.0 text-to-speech task.
About 15.46 credits for 1,000 characters.
doubao_speech.submit_ttsGet Tts
Retrieve a Doubao Speech 2.0 text-to-speech task state and output.
0.06 點數 / 次請求
doubao_speech.get_ttsGenerate Speech
Generate speech with Qwen-Audio TTS.
About 3.66 credits for 1,000 characters with Flash; Plus costs more.
qwen.generate_speechDesign Fish Voice Previews
Generate one to four original voice previews from a description. Returns base64 WAV audio synchronously without creating a persistent voice model. Decode each audio_base64 value to save the preview.
約 3.09 點數 / 次請求
aisa.design_fish_voice_previewsQwen Tts Generate
Generate speech audio
28.8 點數 / 1 萬字起
- Qwen3 Flash
- 28.8 點數 / 1 萬字
- Qwen-Audio 3.0 Flash
- 36 點數 / 1 萬字
- Qwen-Audio 3.0 / 3.1 Plus
- 50.4 點數 / 1 萬字
fusion-api.qwen_tts_generate沒有符合的工具。
常見問題
How should I prepare narration text?
Review names, numbers, punctuation, and sentence length before generation. Include pronunciation guidance when the selected service supports it.