magenta / ddsp

DDSP: Differentiable Digital Signal Processing
https://magenta.tensorflow.org/ddsp
Apache License 2.0
2.91k stars 337 forks source link

AttributeError: module 'hmmlearn.hmm' has no attribute 'CategoricalHMM' #499

Closed RamakanthRGunishetty closed 1 year ago

RamakanthRGunishetty commented 1 year ago

AttributeError Traceback (most recent call last) in <cell line: 30>() 28 # Compute features. 29 start_time = time.time() ---> 30 audio_features = ddsp.training.metrics.compute_audio_features(audio) 31 audio_features['loudness_db'] = audio_features['loudness_db'].astype(np.float32) 32 audio_features_mod = None

3 frames /usr/local/lib/python3.8/dist-packages/ddsp/training/metrics.py in compute_audio_features(audio, n_fft, sample_rate, frame_rate) 67 68 audio_feats['f0_hz'], audio_feats['f0_confidence'] = ( ---> 69 ddsp.spectral_ops.compute_f0(audio, sample_rate, frame_rate)) 70 71 return audio_feats

/usr/local/lib/python3.8/dist-packages/ddsp/spectral_ops.py in compute_f0(audio, sample_rate, framerate, viterbi) 294 295 # Compute f0 with crepe. --> 296 , f0_hz, f0confidence, = crepe.predict( 297 audio, 298 sr=sample_rate,

/usr/local/lib/python3.8/dist-packages/crepe/core.py in predict(audio, sr, model_capacity, viterbi, center, step_size, verbose) 259 260 if viterbi: --> 261 cents = to_viterbi_cents(activation) 262 else: 263 cents = to_local_average_cents(activation)

/usr/local/lib/python3.8/dist-packages/crepe/core.py in to_viterbicents(salience) 142 143 # fix the model parameters because we are not optimizing the model --> 144 model = hmm.CategoricalHMM(360, starting, transition) 145 model.startprob, model.transmat, model.emissionprob = \ 146 starting, transition, emission

AttributeError: module 'hmmlearn.hmm' has no attribute 'CategoricalHMM'