Connect Doubao Speech to Your AI Agent

Choose a voice for narration or extract text from an existing recording.

4 tools API key

Try a Doubao Speech Task

Copy a prompt into your agent with the oo Skill installed. Add your page links and source notes before running it.

oo Skill Doubao Speech

Transcribe and summarize a recording

Recognize a supplied audio file and organize its content with available segment times.

/oo Doubao Speech: Transcribe [public audio URL] in format [format], language [language]. Retrieve the completed transcript and summarize [requested topics], including relevant segment times when available. If still processing, give me the task ID and state.

Generate narration audio

Synthesize supplied narration with a chosen speaker and file format.

/oo Doubao Speech: Generate Doubao Speech narration for [text] with speaker ID [voice ID] in [audio format]. Check the task and provide the completed audio link and its expiry information when available.

Start Working with Doubao Speech

Install OOMOL

Set up oo CLI and the oo Skill in your agent.
Install OOMOL →
Connect Doubao Speech

Connect with a Doubao Speech APP Key, or use these speech operations through Marketplace.
Connect Doubao Speech →
Give Your Agent a Task

Use /oo with the pages to read, the changes to make, and where to save the result.
Connection Guide →

Choose Your Agent

Codex Claude Code Cursor Hermes

For synthesis, provide text and a speaker ID. For recognition, provide a public audio URL and its format.
Ark API keys are not accepted for this connection. Tasks are asynchronous, and synthesized audio links are temporary.

Doubao Speech Tools

Submit text-to-speech and recording recognition tasks, then retrieve their progress and completed outputs. AI can prepare narration text or organize recognized speech into a summary using the transcript and available segment times.

Recording transcription

Submit an audio URL and retrieve recognized text and timed segments.

Speech synthesis

Generate speech from text and retrieve its temporary audio link.

4 tools

Submit Tts

Submit an asynchronous Doubao Speech 2.0 text-to-speech task.

doubao_speech.submit_tts

Get Tts

Retrieve a Doubao Speech 2.0 text-to-speech task state and output.

doubao_speech.get_tts

Submit Stt

Submit an asynchronous Doubao recording-file speech recognition task.

doubao_speech.submit_stt

Get Stt

Retrieve a Doubao recording-file speech recognition task state and output.

doubao_speech.get_stt

Tools are available through the OOMOL marketplace. Usage charges may apply.

Frequently Asked Questions

What do I need for recording recognition?

Provide a publicly accessible audio file URL, its format and a language if needed. Recognition works on a recording file rather than a live microphone session.

What can a completed transcript include?

Recognized text, source duration and segment start and end times when available. Those segments can help locate passages in the recording.

How do I choose the narration voice?

Supply a Doubao Speech speaker ID along with the text and desired audio format. The connection does not expose a voice-listing operation.