Librosa: Filter pass-band beyond Nyquist

Question

Librosa Python 3.5 On 8000Hhz Audio samples getting error:

extract feature error. Filter pass-band lies beyond Nyquist

Was previously getting other errors about Nyquist, but fixed by setting sr and fmin manually, can't seem to get past this one though.

stft = np.abs(librosa.stft(X))
mfccs = np.mean(librosa.feature.mfcc(y=X, sr=8000, n_mfcc=40, fmin=60).T,axis=0)
chroma = np.mean(librosa.feature.chroma_stft(S=stft, sr=8000).T,axis=0)
mel = np.mean(librosa.feature.melspectrogram(X, sr=8000, fmin=60).T,axis=0)
contrast = np.mean(librosa.feature.spectral_contrast(S=stft, sr=8000, fmin=60).T,axis=0)
tonnetz = np.mean(librosa.feature.tonnetz(y=librosa.effects.harmonic(X), sr=8000).T,axis=0)
return mfccs, chroma, mel, contrast, tonnetz

Edit:

Problem seems to exist in the tonnetz feature. Per docs chroma :

np.ndarray [shape=(n_chroma, t)] or None

    Normalized energy for each chroma bin at each frame.

    If None, a cqt chromagram is performed.

A cqt chromagram would require additional parameters to work with 8000Hhz audio, therefor I would need to set chroma manually. Unfortunately this is beyond my Librosa/Audio knowledge, anyone got a clue for me?

score 1 · Answer 1 · answered Jun 05 '19 at 06:59

    stft = np.abs(librosa.stft(X))
    mfccs = np.mean(librosa.feature.mfcc(y=X, sr=8000, n_mfcc=40, fmin=30).T,axis=0)
    chroma = np.mean(librosa.feature.chroma_stft(S=stft, sr=8000).T,axis=0)
    mel = np.mean(librosa.feature.melspectrogram(X, sr=8000, fmin=30).T,axis=0)
    contrast = np.mean(librosa.feature.spectral_contrast(S=stft, sr=8000, fmin=30).T,axis=0)
    tonnetz = np.mean(librosa.feature.tonnetz(y=librosa.effects.harmonic(X), sr=sample_rate, chroma=librosa.feature.chroma_cqt(y=X, sr=8000, fmin=30)).T,axis=0)
    return mfccs, chroma, mel, contrast, tonnetz

Librosa: Filter pass-band beyond Nyquist

1 Answers1