Extract Pitch Class Histograms from music

1. Select media files or record live audio
... or drop your audio files
... or record audio
2. Analyze the Pitch Class Histogram
Min 90 cents apart

Min 10% high

Speed 100%
4. Download results

More information

What is this for?
This page tries to extract pitch class histograms from audio. Pitch class histograms give pointers on how pitch is used in a performance and are especially relevant for music of which the scale is not known: think about, for example, historic recordings of non-eurogenetic music cultures. It also says something about
How do I use this?
Drop the media files you want to analyze the pitch organization from on the placeholder above. The file is automatically analyzed for pitch patterns and a resulting graph is shown. The data, which is extracted during the process can be downloaded seperately.
What is in these files I downloaded?
The pitch class histogram gives tels you how common each pitch (expressed in cents) is within an octave. By convention the pitch class histogram starts on a C (or for example 261.63Hz, midi key 60 or octaves apart). The pitch histogram is a list of 12800 values: each cent value in the full midi range (0-128). Each distance between two keys on a keyboard is split into 100 (which explains the name cents). The pitch estimations are the base information extracted with a pitch extractor called YIN. YIN works best for vocal performances.
What happens if I drop a video file?
The application automatically uses the first audio stream in the video container for pitch extraction.
I do not want to share any audiovisual material. Can I use this page?
There is no media uploaded to the server or shared in any way. Transcoding and synchronization happens at client side thanks to the power of WebAssembly and Javascript.