Audio Fingerprint Identification

The following diagram shows the modules in HPE IDOL Speech Server that create audio fingerprints in a single step.

The wav module reads the audio file and prepares windowed data.

a is the audio window series.

 

The afpfeature module takes each window of samples and converts it to a time series of AFP features, based on significant audio events.

fp is the feature vector series.


The afp module matches the AFP features against tracks stored in the fingerprint database feature vectors.

tr is the recognized track labels and time positions.


The afpout module prepares the output audio track labels and time positions for storage and result reporting.

The schema that implements this feature is:

[MyAfpMatch]
a ← wav (MONO, input)
fp ← afpfeature (_, a)
tr ← afp (_, fp)
output ← afpout (_, tr)

_HP_HTML5_bannerTitle.htm