Music and soundSearch and RAG

AST (Audio Spectrogram Transformer)

A classic 2021 sound recognition model: detects 527 AudioSet event classes (siren, barking, breaking glass, music). Lightweight, runs without a GPU, in Transformers since 2022.

The last open version came out in Nov 2022. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
MIT, USA
First release
Nov 2022
Latest release
Nov 2022
Sizes
about 87M
License
Commercial use allowedBSD 3-Clause
Running
On your own serverAlso runs without a GPU
Industries
Security, Manufacturing and logistics, Media and production
Deploy on your server Hugging Face GitHub

What it does

  • Sound event recognition
  • Tagging an audio archive
  • Detecting alarm sounds
  • Voice command recognition

Where it is used

Security and monitoringManufacturing (acoustic monitoring)Media

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. AST, дообученные на AudioSet и Speech Commands (в Transformers)

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can AST (Audio Spectrogram Transformer) be used in a commercial project?

Yes. License: BSD 3-Clause. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does AST (Audio Spectrogram Transformer) need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does AST (Audio Spectrogram Transformer) support Russian?

Language does not matter for this model: it does not work with text.

Where can I download AST (Audio Spectrogram Transformer) and what does it cost?

The AST (Audio Spectrogram Transformer) weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Music and soundMERTm-a-p (Multimodal Art Projection) · UK / ChinaNon-commercial only

A music encoder: turns a track into a numeric representation used to detect genre, mood, key and rhythm. MERT-v2 handles full songs up to 6 minutes.

DetailsMusic and soundMuQ / MuQ-MuLanTencent AI Lab · ChinaNon-commercial only

The MuQ music encoder and the MuQ-MuLan model, which matches music and text: you can search for tracks by a description in English or Chinese.

DetailsVoice: speakers and soundSAM AudioMeta · USACommercial use with conditions

A model that cuts the sound you need out of a recording based on a text description, a mark on the video or a time range: a voice, an instrument, noise. Weights are available on request.

Details

Source: huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.