github/dejavu
1
0
Fork 0
mirror of https://github.com/correl/dejavu.git synced 2024-12-28 03:00:04 +00:00
Code Issues Projects Releases Packages Wiki Activity

Added tuning section to README.md. Removed trailing space from fingerprint.py

Browse source
This commit is contained in:
Saleem Ansari 2014-11-28 03:58:52 +05:30
parent efe3a33915
commit 4d2b91b052
2 changed files with 24 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
README.md
View file

@ -100,6 +100,19 @@ An example configuration is as follows:
>>> djv = Dejavu(config)
```
## Tuning
Inside `fingerprint.py`, you may want to adjust following parameters (some values are given below).
FINGERPRINT_REDUCTION = 30
PEAK_SORT = False
DEFAULT_OVERLAP_RATIO = 0.4
DEFAULT_FAN_VALUE = 10
DEFAULT_AMP_MIN = 15
PEAK_NEIGHBORHOOD_SIZE = 30
These parameters are described in the `fingerprint.py` in detail. Read that in-order to understand the impact of changing these values.
## Recognizing
There are two ways to recognize audio using Dejavu. You can recognize by reading and processing files on disk, or through your computer's microphone.

22
dejavu/fingerprint.py
View file

@ -12,8 +12,8 @@ IDX_TIME_J = 1
######################################################################
# Sampling rate, related to the Nyquist conditions, which affects
# the range frequencies we can detect.
DEFAULT_FS = 44100
# the range frequencies we can detect.
DEFAULT_FS = 44100
######################################################################
# Size of the FFT window, affects frequency granularity
@ -23,15 +23,15 @@ DEFAULT_WINDOW_SIZE = 4096
# Ratio by which each sequential window overlaps the last and the
# next window. Higher overlap will allow a higher granularity of offset
# matching, but potentially more fingerprints.
DEFAULT_OVERLAP_RATIO = 0.5
DEFAULT_OVERLAP_RATIO = 0.5
######################################################################
# Degree to which a fingerprint can be paired with its neighbors --
# higher will cause more fingerprints, but potentially better accuracy.
DEFAULT_FAN_VALUE = 15
# higher will cause more fingerprints, but potentially better accuracy.
DEFAULT_FAN_VALUE = 15
######################################################################
# Minimum amplitude in spectrogram in order to be considered a peak.
# Minimum amplitude in spectrogram in order to be considered a peak.
# This can be raised to reduce number of fingerprints, but can negatively
# affect accuracy.
DEFAULT_AMP_MIN = 10
@ -39,13 +39,13 @@ DEFAULT_AMP_MIN = 10
######################################################################
# Number of cells around an amplitude peak in the spectrogram in order
# for Dejavu to consider it a spectral peak. Higher values mean less
# fingerprints and faster matching, but can potentially affect accuracy.
# fingerprints and faster matching, but can potentially affect accuracy.
PEAK_NEIGHBORHOOD_SIZE = 20
######################################################################
# Thresholds on how close or far fingerprints can be in time in order
# Thresholds on how close or far fingerprints can be in time in order
# to be paired as a fingerprint. If your max is too low, higher values of
# DEFAULT_FAN_VALUE may not perform as expected.
# DEFAULT_FAN_VALUE may not perform as expected.
MIN_HASH_TIME_DELTA = 0
MAX_HASH_TIME_DELTA = 200
@ -56,7 +56,7 @@ MAX_HASH_TIME_DELTA = 200
PEAK_SORT = True
######################################################################
# Number of bits to throw away from the front of the SHA1 hash in the
# Number of bits to throw away from the front of the SHA1 hash in the
# fingerprint calculation. The more you throw away, the less storage, but
# potentially higher collisions and misclassifications when identifying songs.
FINGERPRINT_REDUCTION = 20
@ -137,7 +137,7 @@ def generate_hashes(peaks, fan_value=DEFAULT_FAN_VALUE):
[(e05b341a9b77a51fd26, 32), ... ]
"""
fingerprinted = set() # to avoid rehashing same pairs
if PEAK_SORT:
peaks.sort(key=itemgetter(1))