FAQ

The basic concept of digital audio watermarking is to insert an inaudible secret message (data, code, identifier) inside the sound source content and remove this information, causing the sound source to lose its function. AD uses its own watermark engine to insert inaudible information into audio content. This information can be used to search for and determine the owners and recipients of the audio content.

Watermark verification is the process of analyzing, detecting, and extracting digital watermark (signature) information from sound source files. This process is performed by the AD decoding engine. Users can get analysis results in a short time by uploading sound source files. When a signature is found, the information is displayed on the screen. The Personal Signature field appears only to the signature owner (i.e., the AD user who created the signature). The Open Signature field can be viewed by all users, including unregistered and unlogged AD visitors.

A signature (author's metadata) is information about the owner and recipient, which is a watermark included in an audio file. Private fields are not publicly visible and can only be read by the original owner who generated the signature. Private fields are mainly used for backtracking and source proofing. Information about the file recipient is part of the signature and is recorded in the file during the preparation phase before the file is passed to the recipient. This is also called forensics watermarking. The public field is the field that the user has checked the public option among the private fields. Fields marked 'Public' at the time of signature generation can be retrieved and read by anyone, including unregistered visitors. For privacy purposes, information about file recipients becomes a private field. If none of the signature fields is selected for publication, no one but you can extract signature data or verify whether the audio is signed or not. If you select some fields (for example, name and audio track information) to be public, anyone with a file can use the AD Verification Tool to analyze sound source content and retrieve public signature data. You can claim the copyright and ownership of the sound source by sharing the sound source with the world including the public signature. Even if the audio file is renamed, converted, and metadata is removed, the sound source content still contains watermarks. So anyone can check the source and ownership of the file. Secondary licensing and related authorization transactions will be available on the content exchange to be established later.

I found audio files that I shared with multiple people just for reference or for monitoring were published and circulating on the Internet. We don't know who leaked it. However, if you have sent a file with unique messages and recipient information for each recipient, you can track the source of the leak immediately. That's how Audio Defence works.

After uploading the file, you can insert your own watermark for each recipient and 'send' it. A unique watermark copy of the uploaded file is sent to the recipient immediately by email (download link with expiration date) or you can download a copy with the watermark directly and then share it with the recipient manually. In this case, there is no record of file transmission in the server other than the information owned by the person, so when a problem occurs, the personal transmission record must be checked directly, and the watermark of the leaked file is the same, making it difficult to find the responsibility of the leaked person. Therefore, it is highly recommended to insert and share different watermarks for each destination and use of the same sound source. For frequently used files, you can upload them to AD File Storage and save them. Using the same sharing method, you can share files by issuing new watermarks directly from Storage without having to upload files repeatedly each time.

Watermarks and audio are protected indefinitely. Your personal signature will remain extractable as long as your account is active. Public signatures remain extractable as long as Audio Defence exists.

No.

AD watermarks are resistant to conversion attacks such as sample rate conversion, A/D, D/A conversion, MP3, Ogg Borbis, and other lossy audio codecs, noise addition, editing, EQ and compression, limitation, and distortion, and it is practically impossible to remove watermarks without the physical loss of the original sound quality.

Audio Defence service does not support direct encoding of video files. However, it is possible for the user to encode the video after inserting a watermark into the audio to be included in the video.

AD only provides technical means to trace and find illegal use. However, for users who need legal assistance regarding copyright, we provide a link to get help from a copyright lawyer, and only Korean regions are available for now.