Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > ComfyUI-API-Toolkit > ElevenLabs - Sound Effects

ComfyUI Node: ElevenLabs - Sound Effects

Class Name

AIS_EL_SFX

Category
API Toolkit/ElevenLabs/Audio
Author
ixmxamar (Account age: 0days)
Extension
ComfyUI-API-Toolkit
Latest Updated
2026-10-02
Github Stars
0.0K

How to Install ComfyUI-API-Toolkit

Install this extension via the ComfyUI Manager by searching for ComfyUI-API-Toolkit
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-API-Toolkit in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

ElevenLabs - Sound Effects Description

Generates sound effects from text using ElevenLabs and outputs audio for ComfyUI workflows.

ElevenLabs - Sound Effects:

AIS_EL_SFX generates a sound effect from a written description using ElevenLabs sound generation. You can describe an ambience, an impact, or another sound you need for an artwork or scene, then use the generated audio in the rest of your workflow. The node sends the description to the sound-generation service and converts the returned audio into a format that connected audio nodes can use.

ElevenLabs - Sound Effects Input Parameters:

api_key

A required STRING containing the API key used to access the sound-generation service. The field is marked as a password field, so the key is concealed while you enter it. Its default is an empty string; no minimum or maximum length is specified here. The node passes the value to its API-key helper before making a request, but the provided source does not show whether that helper can obtain a key from another location. If authentication fails, check that the service has a valid key available.

text

A required, multiline STRING describing the sound you want to create. Its default is an empty string, and no length limit is specified in this node. The description is sent to the service as the generation prompt, so concrete details such as the sound source, setting, and desired character help communicate your intent. Blank or whitespace-only text stops execution with Text description is empty.

model

An optional COMBO[STRING] whose default is eleven_sfx_v2. Its available choices come from SFX_MODELS, but that list is not included in the supplied context. Although the field is labeled as a sound-effects model selector, the shown generate method does not put its value in the request. Changing this setting therefore has no demonstrated effect on generation in this implementation.

duration

An optional FLOAT specifying the requested length in seconds. Its default is 5.0, minimum is 0.5, maximum is 30.0, and interface step is 0.1. When auto_duration is off, the node sends this value as duration_seconds; choose a shorter length for a brief sound or a longer one when the effect needs room to develop. When auto_duration is on, the node does not send duration.

prompt_influence

An optional FLOAT controlling how closely generation follows your description. Its default is 0.3, minimum is 0.0, maximum is 1.0, and interface step is 0.01. The node sends the value with every generation request. A higher setting asks for closer adherence to the prompt; a lower setting allows more freedom, though the exact audible difference depends on the service’s response.

loop

An optional BOOLEAN with a default of False. Enable it to request a smoothly looping sound effect, such as an ambience intended to repeat. The node sends loop: true only when this setting is enabled; otherwise, it leaves the loop setting out of the request. Whether the result loops seamlessly should be checked by listening to its transition from end to beginning.

output_format

An optional COMBO[STRING] selecting the requested audio format. Its default is mp3_44100_128. The available choices come from OUTPUT_FORMATS, which is not shown in the supplied context, so no complete options list can be stated here. The node sends your choice to the service and uses that same choice when converting the response into workflow audio; select a format compatible with the audio nodes you plan to use next.

seed

An optional INT with a default of 0, minimum of 0, and maximum of 4294967295. A value above 0 is sent with the request and can be used when you want to repeat a particular generation setup. At 0, the node omits the seed from the request. The provided source does not guarantee identical audio across repeated requests, even when you reuse a seed.

auto_duration

An optional BOOLEAN with a default of False. When enabled, the node lets the sound-generation service choose a length based on your description by omitting duration_seconds from the request. Use it when the effect does not need a fixed requested length. When it is enabled, changing duration has no effect on the request.

ElevenLabs - Sound Effects Output Parameters:

audio

An AUDIO output containing the generated sound effect, converted from the service’s returned audio bytes for use by connected audio nodes. This is the node’s only output. Listen to it to judge whether the sound, length, and any requested loop suit your scene; it is audio content, not a text description or a file path.

ElevenLabs - Sound Effects Usage Tips:

  • Describe what you want to hear rather than only naming an object. For example, A short metallic door slam with a hollow echo gives the generator more direction than door.
  • Leave auto_duration off and adjust duration when you have a target length. Turn auto_duration on when you would rather let the service choose; in that mode, the duration setting is ignored.
  • For a repeating background sound, enable loop and listen across the point where playback returns to the start. For a one-time impact, leave it off.
  • Start with the default prompt_influence of 0.3, then change one setting at a time while comparing results. Increase it when you want to ask for closer adherence to your description.
  • Use a seed above 0 when comparing prompt or setting changes and you want to send a specific seed with each request. Use 0 when you do not need to specify one.
  • Do not rely on changing model to alter the result in this version: the provided method accepts that input but does not send it to the sound-generation service.

ElevenLabs - Sound Effects Common Errors and Solutions:

Text description is empty.

  • Explanation: The node raises this exact error when text is blank or contains only whitespace, before it makes a generation request.
  • Solution: Enter a description of the sound effect in text and run the node again.

Authentication or invalid API key error

  • Explanation: A request may fail if the service does not receive a valid API key. The exact message depends on the API-key helper or the service response; neither error string is shown in the provided source.
  • Solution: Check that your key is valid and available to the node, then retry. Keep the key private.

Access, account, or usage-limit error

  • Explanation: The service may reject a request because the account cannot use the requested service or has reached an applicable limit. The node checks the response, but the supplied source does not establish a particular message.
  • Solution: Review the account’s access and usage information, resolve any limit or permission issue shown by the service, and retry.

Invalid request or unsupported output format error

  • Explanation: The service may reject a setting it does not accept. The node sends text and prompt_influence, optionally sends duration_seconds, loop, and seed, and passes output_format as a request parameter. The precise error text is service-dependent.
  • Solution: Restore the displayed defaults, confirm that output_format is one of the available choices, and retry. Reintroduce changed settings one at a time to identify the cause.

Request timed out

  • Explanation: The node sets a 300-second timeout for the generation request. If a response does not arrive in time, the request can fail; the exact timeout message depends on the API client.
  • Solution: Check your connection and the service’s availability, then retry. You can also try a simpler description or a shorter requested duration with auto_duration off.

Audio conversion or decoding error

  • Explanation: After a successful response check, the node converts the returned bytes using the selected output_format. If those bytes cannot be converted, generation will not produce a usable AUDIO output; the provided source does not show an exact error message.
  • Solution: Retry with the default output_format of mp3_44100_128. If the problem persists, check whether the service returned valid audio and whether your environment supports that format.

ElevenLabs - Sound Effects Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-API-Toolkit
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

ElevenLabs - Sound Effects