This repo contains the Inference code for AudioSeal, a method for speech localized watermarking, with state-of-the-art robustness and detector speed (training code coming soon).
To learn more, check out our paper.
[arXiv]
[🤗Hugging Face]
[Colab Notebook]
[Webpage]
[Blog]
[Press]
- 2024-06-17: Training code is now available. Check the instruction!!!
- 2024-05-31: Our paper gets accepted at ICML'24 :)
- 2024-04-02: We have updated our license to full MIT license (including the license for the model weights) ! Now you can use AudioSeal in commercial application too!
- 2024-02-29: AudioSeal 0.1.2 is out, with more bug fixes for resampled audios and updated notebooks
AudioSeal introduces a breakthrough in proactive, localized watermarking for speech. It jointly trains two components: a generator that embeds an imperceptible watermark into audio and a detector that identifies watermark fragments in long or edited audio files.
- Key Features:
- Localized watermarking at the sample level (1/16,000 of a second).
- Minimal impact on audio quality.
- Robust against various audio edits like compression, re-encoding, and noise addition.
- Fast, single-pass detection designed to surpass existing models significantly in speed — achieving detection up to two orders of magnitude faster, making it ideal for large-scale and real-time applications.
pip install audioseal
To install from source: Clone this repo and install in editable mode:
git clone https://github.com/facebookresearch/audioseal
cd audioseal
pip install -e .
You can find all the model checkpoints on the Hugging Face Hub. We provide the checkpoints for the following models:
- AudioSeal Generator: Takes an audio signal (as a waveform) and outputs a watermark of the same size as the input, which can be added to the input to watermark it. Optionally, it can also take a secret 16-bit message to embed in the watermark.
- AudioSeal Detector: Takes an audio signal (as a waveform) and outputs the probability that the input contains a watermark at each sample (every 1/16k second). Optionally, it may also output the secret message encoded in the watermark.
Note that the message is optional and has no influence on the detection output. It may be used to identify a model version for instance (up to
Here’s a quick example of how you can use AudioSeal’s API to embed and detect watermarks:
from audioseal import AudioSeal
# model name corresponds to the YAML card file name found in audioseal/cards
model = AudioSeal.load_generator("audioseal_wm_16bits")
# Other way is to load directly from the checkpoint
# model = Watermarker.from_pretrained(checkpoint_path, device = wav.device)
# a torch tensor of shape (batch, channels, samples) and a sample rate
# It is important to process the audio to the same sample rate as the model
# expects. In our case, we support 16khz audio
wav, sr = ..., 16000
watermark = model.get_watermark(wav, sr)
# Optional: you can add a 16-bit message to embed in the watermark
# msg = torch.randint(0, 2, (wav.shape(0), model.msg_processor.nbits), device=wav.device)
# watermark = model.get_watermark(wav, message = msg)
watermarked_audio = wav + watermark
detector = AudioSeal.load_detector("audioseal_detector_16bits")
# To detect the messages in the high-level.
result, message = detector.detect_watermark(watermarked_audio, sr)
print(result) # result is a float number indicating the probability of the audio being watermarked,
print(message) # message is a binary vector of 16 bits
# To detect the messages in the low-level.
result, message = detector(watermarked_audio, sr)
# result is a tensor of size batch x 2 x frames, indicating the probability (positive and negative) of watermarking for each frame
# A watermarked audio should have result[:, 1, :] > 0.5
print(result[:, 1 , :])
# Message is a tensor of size batch x 16, indicating of the probability of each bit to be 1.
# message will be a random tensor if the detector detects no watermarking from the audio
print(message) Interested in training your own watermarking model? Check out our training documentation to get started.
We welcome pull requests with improvements or suggestions. If you wish to report an issue or propose an enhancement but are unsure how to implement it, feel free to create a GitHub issue.
-
If you encounter the error
ValueError: not enough values to unpack (expected 3, got 2), this is because we expect a batch of audio tensors as inputs. Add one dummy batch dimension to your input (e.g.wav.unsqueeze(0), see example notebook for getting started). -
In Windows machines, if you encounter the error
KeyError raised while resolving interpolation: "Environmen variable 'USER' not found": This is due to an old checkpoint uploaded to the model hub, which is not compatible in Windows. Try to invalidate the cache by removing the files inC:\Users\<USER>\.cache\audiosealand re-run again. -
If you use torchaudio to handle your audios and encounter the error
Couldn't find appropriate backend to handle uri ..., this is due to newer version of torchaudio does not handle the default backend well. Either downgrade your torchaudio to2.1.0or earlier, or installsoundfileas your audio backend.
- The code in this repository is licensed under the MIT license as detailed in the LICENSE file. This license permits reuse, modification, and distribution of the software, as long as the original license is included.
If you find this repository useful, please consider giving it a star ⭐ and citing our work:
@article{sanroman2024proactive,
title={Proactive Detection of Voice Cloning with Localized Watermarking},
author={San Roman, Robin and Fernandez, Pierre and Elsahar, Hady and D´efossez, Alexandre and Furon, Teddy and Tran, Tuan},
journal={ICML},
year={2024}
}