AudioTag.info — is a free music-recognition service. It allows you to identify almost any unknown piece of music quickly and easily. Its use is very simple: you upload a short audio fragment (even 15 seconds are enough) or an entire song, the robot analyzes it and provides you with the information about the track title, artist name, album title, etc. Your audio fragment can be in almost any file format and of almost any quality (aurally recognizable, of course) — it can be an MP3 file downloaded from the Internet or a short recording made with your old tape recorder and stored as a low-quality .WAV-file.
Yes, the service is absolutely free.
This service is based on a sophisticated audio recognition algorithm combining advanced audio fingerprinting technology and a large songs' database. When you upload an audio file, it is being analyzed by an audio engine. During the analysis its audio “fingerprint” is extracted and identified by comparing it to the music database. At the completion of this recognition process, information about songs with their matching probabilities are displayed on screen.
You can upload your files in almost any file format and with any bit-rate. Uncompressed, ADPCM and other waves (.WAV), MPEG-1 Layer III (.MP3), Ogg Vorbis (.OGG), FLAC, Flash Video (.FLV), .AMR, .MP4, and many more file formats are supported.
Yes, the recognition is reliable. Of course, erroneous recognition is possible, but its probability is low. The algorithm is designed so that it will rather not identify a song than show an incorrect result.
An optimal duration of an analyzed (uploaded) piece of an audio is 15-45 seconds. Shorter and longer audio files can be recognized as well. Generally, recognition process is performed as follows: uploaded file is sliced into portions of 15 seconds each; 1 to 3 music slices are chosen at random and analyzed one by one. Therefore, to save your time and bandwidth, it is recommended to upload relatively short audio fragments of 15 to 45 seconds.
When you choose a song fragment for uploading, please make sure it is indicative, i.e. that it contains enough musical and audio information to distinguish it. Generally, you should take some slice from the middle of the song. If you are not sure - upload the whole song and the recognition engine will choose slices randomly by itself.
The algorithm analyses the acoustic characteristics of sound and does not consider any melodic and/or musical properties. Therefore, it may be unable to find only melodically similar tracks and can only find tracks with an exactly matching sound. Of course, it does not mean the algorithm cannot recognize noisy/distorted recordings.
The songs database counts more than one million tracks
The database contains tracks with all kinds of music from modern popular music to old jazz and classics. There are also tracks with Russian music.
Unfortunately, we do not support user submissions to the database at this stage. But we plan to implement this service in the future.
This service does not compete with others. Some of the main features of this service are the ability to recognize audio even by a very short piece of music, and its online availability does not require you to install any software on your computer. Anyway, it is another tool helping you to identify audio, and it is quite precise.