Skip to content
SPECIFICTION
Get listed

AudioGen

github.com

Free plan LinuxAPISelf-hosted
Visit github.com
Spec score 8.8 #2 of 29 in AI Sound Effect Generators

What it is

AudioGen is a text-to-sound model provided through AudioCraft. It generates audio samples from text descriptions and can continue from an audio prompt, with conditional and unconditional generation modes. Sampling controls include greedy, temperature, top-K, and top-P methods. The listed pretrained model, facebook/audiogen-medium, has 1.5 billion parameters; the example API generates sound and saves it as WAV. The implementation uses an autoregressive Transformer and a 16 kHz EnCodec tokenizer. The repository provides API access and a local Jupyter notebook demo. Running the medium model locally requires a GPU with at least 16 GB of memory; installation requires Python 3.9 and PyTorch 2.1.0, with ffmpeg recommended. The model was trained on English descriptions and performs less well in other languages. It cannot generate realistic vocals, and prompts may need refinement. Code is MIT-licensed, while weights use CC-BY-NC 4.0; training datasets are not provided. The listed self-hosted plan is free.

Spec sheet

What it does
AudioGen is a text-guided audio generation model that generates sounds from text.Read 3 Oct 2026 · github.com
Purpose
AudioGen is a text-to-sound generation model provided through AudioCraft.Read 3 Oct 2026 · github.com
Intended users
The model card names audio, machine learning, and AI researchers, as well as amateurs learning about generative models, as primary users.Read 3 Oct 2026 · github.com
Free plan
YesRead 20 Sep 2026 · github.com
Support
The model card directs questions and comments to the project’s GitHub repository or its issue tracker.Read 3 Oct 2026 · github.com
License
The repository says its code is MIT-licensed and its model weights use CC-BY-NC 4.0.Read 3 Oct 2026 · github.com
Audio continuation
The generation stage supports conditional or unconditional sample generation and audio continuation from a prompt.Read 3 Oct 2026 · github.com
Available model
The page lists one pretrained AudioGen model, facebook/audiogen-medium, with 1.5 billion parameters.Read 3 Oct 2026 · github.com
Generation
The example API generates sound from text descriptions and shows saving output as WAV audio.Read 3 Oct 2026 · github.com
Generation modes
The training documentation describes conditional and unconditional generation, audio continuation from a prompt, and greedy, temperature, top-K, and top-P sampling.Read 3 Oct 2026 · github.com
Hardware requirement
The instructions say inference with the medium-sized models requires a GPU with at least 16 GB of memory.Read 3 Oct 2026 · github.com
Installation
AudioCraft installation requires Python 3.9 and PyTorch 2.1.0; ffmpeg is also recommended by the repository instructions.Read 3 Oct 2026 · github.com
Language limit
The model card says AudioGen was trained on English descriptions and performs less well in other languages.Read 3 Oct 2026 · github.com
Local demo
The maker provides a Jupyter notebook demo that can be run locally with a GPU.Read 3 Oct 2026 · github.com
Model architecture
The released model combines EnCodec audio tokenization with an autoregressive Transformer language model and has 1.5 billion parameters.Read 3 Oct 2026 · github.com
Model availability
The AudioGen instructions list one pretrained model, facebook/audiogen-medium, and a local Jupyter notebook demo.Read 3 Oct 2026 · github.com
Model design
The provided reimplementation is a single-stage autoregressive Transformer trained over a 16 kHz EnCodec tokenizer with four codebooks sampled at 50 Hz.Read 3 Oct 2026 · github.com
Model distinction
The provided models are not the original models used to report results in the AudioGen publication.Read 3 Oct 2026 · github.com
Output limit
The model card says AudioGen cannot generate realistic vocals and may require prompt engineering for satisfying results.Read 3 Oct 2026 · github.com
Prompt generation
The documented API generates audio samples from text descriptions, with an example configured to generate five-second samples.Read 3 Oct 2026 · github.com
Responsible use
The model card advises against downstream use without further risk evaluation and mitigation.Read 3 Oct 2026 · github.com
Sampling controls
Generation supports greedy sampling, temperature sampling, top-K sampling, and top-P nucleus sampling.Read 3 Oct 2026 · github.com
Training
AudioGenSolver implements the training pipeline, but the maker says it may not fully reproduce the paper results and does not provide the AudioGen training datasets.Read 3 Oct 2026 · github.com
Training data
The instructions say the datasets used to train AudioGen are not provided.Read 3 Oct 2026 · github.com

Plans and prices

AudioGen (self-hosted) Free No price stated by maker Pretrained medium model · local inference requires a GPU with at least 16 GB memory Read 3 Oct 2026 · github.com

Every AudioGen plan and price →

Strong points and limits

Strong points

  • Generates sound from text and supports audio continuation.
  • Offers greedy, temperature, top-K, and top-P sampling.
  • API example saves generated audio as WAV.
  • Code is MIT-licensed.

Limits

  • Local medium-model inference requires at least 16 GB GPU memory.
  • Trained on English descriptions; other languages perform less well.
  • Cannot generate realistic vocals.
  • Training datasets are not provided.

Who it suits

AudioGen suits audio, machine learning, and AI researchers, as well as amateurs learning about generative models. It may fit users who can work with text prompts and, for local inference, have the specified GPU memory.

Verdict

AudioGen offers prompt-based sound generation with several generation modes and sampling controls. Its language, vocal-generation, hardware, and dataset limits matter when assessing whether it fits a project.

Where Specifiction places it

Alternatives to AudioGenAll alternatives →