Name: ast finetuned audioset 10 10 0.4593
Author: MIT

Overview

This model is a fine-tuned implementation of the Audio Spectrogram Transformer (AST), specifically optimized for the AudioSet dataset. Unlike traditional CNN-based audio models, it treats audio spectrograms as sequences of patches, leveraging a transformer architecture to capture global dependencies across the frequency and time domains. For developers, this means superior performance in multi-label audio classification and event detection. It is particularly effective for building environmental sound recognition systems, automated tagging for media libraries, or triggering actions based on specific acoustic signatures. Integration is straightforward for those familiar with PyTorch or Hugging Face, as it follows standard transformer input pipelines. Compared to baseline models, this version offers a refined balance between accuracy and inference speed for complex audio scenes.

Highlights

Transformer-based architecture for superior audio event detection
Fine-tuned on AudioSet for diverse sound classification
Efficient processing of audio spectrogram patches
BSD-3-Clause license allows flexible commercial integration
High accuracy in multi-label acoustic scene analysis

Usage

Install

# Install Hugging Face transformers
pip install transformers torch

SDK Usage

# Load model with transformers
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("MIT/ast-finetuned-audioset-10-10-0.4593")
tokenizer = AutoTokenizer.from_pretrained("MIT/ast-finetuned-audioset-10-10-0.4593")

Hugging Face Download

We recommend downloading the model via the Hugging Face CLI or Hub SDK.

Guidance：Before downloading, install huggingface_hub with:

Guidance

pip install -U huggingface_hub

CLI Download

Download the full repository

huggingface-cli download MIT/ast-finetuned-audioset-10-10-0.4593

Download a single file to a local folder (e.g. config.json into ./dir)

huggingface-cli download MIT/ast-finetuned-audioset-10-10-0.4593 config.json --local-dir ./dir

See the official docs for more CLI options

SDK Download

# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('MIT/ast-finetuned-audioset-10-10-0.4593')

Git Download

Make sure git-lfs is installed first

Git Download

git lfs install
git clone https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593

To skip LFS large-file downloads, use:

Skip LFS

GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593

Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.

PyTorch / Transformers Usage

Install Transformers

pip install -U transformers torch

Load the model and run inference

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('MIT/ast-finetuned-audioset-10-10-0.4593')
tokenizer = AutoTokenizer.from_pretrained('MIT/ast-finetuned-audioset-10-10-0.4593')

Full Documentation

来源: HuggingFace

---
license: bsd-3-clause
tags:

audio-classification

---

Audio Spectrogram Transformer (fine-tuned on AudioSet)

Audio Spectrogram Transformer (AST) model fine-tuned on AudioSet. It was introduced in the paper AST: Audio Spectrogram Transformer by Gong et al. and first released in this repository.

Disclaimer: The team releasing Audio Spectrogram Transformer did not write a model card for this model so this model card has been written by the Hugging Face team.

Model description

The Audio Spectrogram Transformer is equivalent to ViT, but applied on audio. Audio is first turned into an image (as a spectrogram), after which a Vision Transformer is applied. The model gets state-of-the-art results on several audio classification benchmarks.

Usage

You can use the raw model for classifying audio into one of the AudioSet classes. See the documentation for more info.