# Suno Speech

> Generate performed spoken audio with Suno Speech, from a one-line brief that Suno scripts itself or from your own script, with tone, voice and background music.

Source: https://nodaro.ai/docs/nodes/audio/suno-speech

The **Suno Speech** node generates spoken audio with Suno's expressive speech. Describe a scene in one line and Suno writes the script and performs it, or give it your own script with a tone, a voice and optional background music. Every run returns two takes to choose from.

## When to use it
- You need a voiceover or narration for a video or a slideshow.
- You want a news intro, an announcement or a podcast opener.
- You want short dialogue lines or character readings to feed [Lip Sync](https://nodaro.ai/docs/nodes/video/lip-sync) or [Merge Video & Audio](https://nodaro.ai/docs/nodes/video/merge-video-audio).
- You want a performed delivery with emotion, or speech over background music, rather than one fixed voice.

For a fixed voice that you choose from a library, use [Text to Speech](https://nodaro.ai/docs/nodes/audio/text-to-speech) instead.

## Quick start
### Add the node

Press Tab on the canvas, or open the node picker in the sidebar, and choose **Audio › Music › Suno Speech**.

### Describe the scene

Leave **Mode** on **Brief** and write one line in **Brief**, for example `A short morning news intro about sunny weather`.

### Run it

Click **Run**. The two takes appear on the node.

### Choose a take

Click the button that shows the number of takes, on the node. Then choose a take in the strip under the node. The **Audio** output follows the take you choose.

Workflow: A script from a Text node is performed by Suno Speech, and the take is put on a video.

- Text → Suno Speech (prompt)
- Suno Speech → Merge Video & Audio (audio)

## Inputs
| Input | Accepts | What it does |
| --- | --- | --- |
| **Prompt** | Text nodes, such as [Text](https://nodaro.ai/docs/nodes/automate/text), [Generate Text - Any LLM](https://nodaro.ai/docs/nodes/automate/prompt) and [Suno Lyrics](https://nodaro.ai/docs/nodes/audio/suno-lyrics) | Replaces the text of the current mode: the brief in Brief mode, the script in Script mode. |

## Outputs
| Output | What it carries |
| --- | --- |
| **Audio** | The selected take. Both takes of every run stay on the node, and the output follows the take you choose. |

## Settings
| Setting | Mode | What it does |
| --- | --- | --- |
| **Mode** | Both | **Brief** (the default): Suno writes the script from your brief and performs it. **Script**: Suno speaks your script. |
| **Brief** | Brief | A one-line description of the scene, up to 3,000 characters. |
| **Script** | Script | The words to speak, up to 5,000 characters. Line breaks are kept. |
| **Tone (optional)** | Script | Tone, rhythm, emotion and setting, up to 1,000 characters. An accent or the style of the music can go here too. |
| **Voice** | Script | **Model picks** (the default), **Female** or **Male**. |
| **Variety** | Script | How much the two takes differ: **Model picks** (the default), **0 — exact style**, **1 — balanced**, **2 — distinct styles**, **3 — bold exploration** or **4 — unreasonably varied**. |
| **Background music** | Script | Adds background music under the speech. It is off by default. |

The limits count the characters after leading and trailing spaces are removed. An empty brief or script, or one over its limit, is refused before the run, and nothing is charged. A setting of the other mode is never sent: a Brief run sends only the brief, and a Script run sends only the script and the options you set.

## Credits
A run costs one fixed price, for both takes, whatever the mode or the settings. The **Run** button on the node and the settings panel show the price before the run. Self-hosted Community and Business editions use no credits.

## Tips
- **Start with a brief.** Hear how Suno reads a scene in Brief mode, then switch to Script mode to fix the words.
- **Put the delivery in Tone.** Write `warm, calm narrator, slightly smiling` in **Tone (optional)**, not in the script. Every word in the script is spoken.
- **Set Variety for the job.** A low **Variety** gives two takes that are almost the same. A high one explores different readings.
- **Check a wired script.** A script from Suno Lyrics or Generate Text runs as it arrives, so check its length first.

## Troubleshooting
- **The run is refused before it starts.** The brief or the script is empty or over its limit. Nothing is charged. Shorten the text, or check the text that is wired into **Prompt**.
- **Suno Extend, Suno Separate or Suno Convert WAV refuses a take.** Suno Speech is a separate model from Suno's music, so its takes cannot be extended, separated or converted to WAV by their Suno ids. The run is refused before it starts, and nothing is charged. [Suno Cover](https://nodaro.ai/docs/nodes/audio/suno-cover) and [Suno Separate](https://nodaro.ai/docs/nodes/audio/suno-separate) accept the take's audio as a file, like any other audio.

## From the API
Send `POST /v1/suno-speech` with `{ "mode": "brief", "prompt": "…" }`, or with `{ "mode": "script", "script": "…", "tone": "…", "variety": 1, "vocalGender": "female", "backgroundMusic": true }`. The answer is `{ jobId }`. The finished job's output has `audioUrl` (the first take), `audioUrls` (both takes) and `sunoTracks` (the id, the title and the duration of each take). `GET /v1/credits/model-cost?model=suno-speech` answers the price.

The MCP tool is `suno_speech`, with the same fields, written `vocal_gender` and `background_music`. See the [MCP tools](https://nodaro.ai/docs/mcp/tools/music#suno_speech).

## Frequently asked questions

### What does Suno Speech do?

Suno Speech performs spoken audio with tone, rhythm and emotion, and can add background music. In Brief mode, Suno writes the script from a one-line description and performs it. In Script mode, it speaks your own words. Every run returns two takes.

### How is Suno Speech different from Text to Speech?

Text to Speech reads your text with one fixed ElevenLabs voice. Suno Speech performs the words, so the delivery changes from take to take, and it can write the script for you and add background music.

### How many credits does Suno Speech cost?

Suno Speech costs one fixed price per run, for both takes, whatever the mode or the settings. The Run button on the node shows the price before the run.

### How long can the script be?

In Script mode, the script can be up to 5,000 characters, and the tone up to 1,000. In Brief mode, the brief can be up to 3,000 characters. A longer text is refused before the run, and nothing is charged.

### Can I extend or separate a Suno Speech take like a song?

Not by its Suno ids. A take is speech, not a song, so Suno Extend, Suno Separate and Suno Convert WAV refuse it by id. Suno Cover and Suno Separate accept the take's audio as a file.
