def test(model_path): test_dataset = SpeakerDatasetTIMITPreprocessed() test_loader = DataLoader(test_dataset, batch_size=hp.test.N, shuffle=True, num_workers=hp.test.num_workers, drop_last=True) embedder_net = SpeechEmbedder() embedder_net.load_state_dict(torch.load(model_path)) embedder_net.eval() avg_EER = 0 for e in range(hp.test.epochs): batch_avg_EER = 0 for batch_id, mel_db_batch in enumerate(test_loader): assert hp.test.M % 2 == 0 enrollment_batch, verification_batch = torch.split( mel_db_batch, int(mel_db_batch.size(1) / 2), dim=1) enrollment_batch = torch.reshape( enrollment_batch, (hp.test.N * hp.test.M // 2, enrollment_batch.size(2), enrollment_batch.size(3))) verification_batch = torch.reshape( verification_batch, (hp.test.N * hp.test.M // 2, verification_batch.size(2), verification_batch.size(3))) perm = random.sample(range(0, verification_batch.size(0)), verification_batch.size(0)) unperm = list(perm) for i, j in enumerate(perm): unperm[j] = i verification_batch = verification_batch[perm] enrollment_embeddings = embedder_net(enrollment_batch) verification_embeddings = embedder_net(verification_batch) verification_embeddings = verification_embeddings[unperm] enrollment_embeddings = torch.reshape( enrollment_embeddings, (hp.test.N, hp.test.M // 2, enrollment_embeddings.size(1))) verification_embeddings = torch.reshape( verification_embeddings, (hp.test.N, hp.test.M // 2, verification_embeddings.size(1))) enrollment_centroids = get_centroids(enrollment_embeddings) sim_matrix = get_cossim(verification_embeddings, enrollment_centroids) # calculating EER diff = 1 EER = 0 EER_thresh = 0 EER_FAR = 0 EER_FRR = 0 for thres in [0.01 * i + 0.3 for i in range(70)]: sim_matrix_thresh = sim_matrix > thres FAR = (sum([ sim_matrix_thresh[i].float().sum() - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (hp.test.N - 1.0) / (float(hp.test.M / 2)) / hp.test.N) FRR = (sum([ hp.test.M / 2 - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (float(hp.test.M / 2)) / hp.test.N) # Save threshold when FAR = FRR (=EER) if diff > abs(FAR - FRR): diff = abs(FAR - FRR) EER = (FAR + FRR) / 2 EER_thresh = thres EER_FAR = FAR EER_FRR = FRR batch_avg_EER += EER print("\nEER : %0.2f (thres:%0.2f, FAR:%0.2f, FRR:%0.2f)" % (EER, EER_thresh, EER_FAR, EER_FRR)) avg_EER += batch_avg_EER / (batch_id + 1) avg_EER = avg_EER / hp.test.epochs print("\n EER across {0} epochs: {1:.4f}".format(hp.test.epochs, avg_EER))
verification_batch = verification_batch[perm] print(enrollment_batch.size()) enrollment_embeddings = embedder_net(enrollment_batch) verification_embeddings = embedder_net(verification_batch) verification_embeddings = verification_embeddings[unperm] enrollment_embeddings = torch.reshape( enrollment_embeddings, (hp.test.N, hp.test.M // 2, enrollment_embeddings.size(1))) verification_embeddings = torch.reshape( verification_embeddings, (hp.test.N, hp.test.M // 2, verification_embeddings.size(1))) enrollment_centroids = get_centroids(enrollment_embeddings) sim_matrix = get_cossim(verification_embeddings, enrollment_centroids) # calculating EER diff = 1 EER = 0 EER_thresh = 0 EER_FAR = 0 EER_FRR = 0 for thres in [0.01 * i + 0.5 for i in range(50)]: sim_matrix_thresh = sim_matrix > thres FAR = (sum([ sim_matrix_thresh[i].float().sum() - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N))
def speaker_verify(npy_file, wav_file_path): utterances_spec = [] #utter_path = wav_file_path #os.path.join(hp.integration.verify_upload_folder, wav_file_path) # path of each utterance utter, sr = librosa.core.load(wav_file_path, hp.data.sr) # load utterance audio # utter, sr = librosa.core.load(wav_file_path, sr=None) # utter, sr = librosa.core.load(wav_file_path) #intervals = librosa.effects.split(utter, top_db=30) # voice activity detection intervals = librosa.effects.split(utter) # this works fine for timit but if you get array of shape 0 for any other audio change value of top_db # for vctk dataset use top_db=100 for interval in intervals: if (interval[1] - interval[0] ) > utter_min_len: # If partial utterance is sufficient long, utter_part = utter[ interval[0]: interval[1]] # save first and last 180 frames of spectrogram. S = librosa.core.stft(y=utter_part, n_fft=hp.data.nfft, win_length=int(hp.data.window * sr), hop_length=int(hp.data.hop * sr)) S = np.abs(S)**2 mel_basis = librosa.filters.mel(sr=hp.data.sr, n_fft=hp.data.nfft, n_mels=hp.data.nmels) # mel_basis = librosa.filters.mel(22050, n_fft=hp.data.nfft, n_mels=hp.data.nmels) S = np.log10(np.dot(mel_basis, S) + 1e-6) # log mel spectrogram of utterances utterances_spec.append(S[:, :hp.data.tisv_frame] ) # first 180 frames of partial utterance utterances_spec.append(S[:, -hp.data.tisv_frame:] ) # last 180 frames of partial utterance if len(utterances_spec) == 0: # no qualified interval found in this audio return -1 wav_file_npy = np.array(utterances_spec) #print("\n############ "+npy_file) npy_file = np.load(npy_file) if shuffle: utter_index = np.random.randint( 0, wav_file_npy.shape[0], utter_num) # select M utterances per speaker wav_file_npy = wav_file_npy[utter_index] utter_index = np.random.randint( 0, npy_file.shape[0], utter_num) # select M utterances per speaker npy_file = npy_file[utter_index] else: wav_file_npy = wav_file_npy[ utter_start:utter_start + utter_num] # utterances of a speaker [batch(M), n_mels, frames] npy_file = npy_file[ utter_start:utter_start + utter_num] # utterances of a speaker [batch(M), n_mels, frames] wav_file_npy = wav_file_npy[:, :, : 160] # TODO implement variable length batch size wav_file_npy = torch.tensor(np.transpose( wav_file_npy, axes=(0, 2, 1))) # transpose [batch, frames, n_mels] npy_file = npy_file[:, :, : 160] # TODO implement variable length batch size npy_file = torch.tensor(np.transpose( npy_file, axes=(0, 2, 1))) # transpose [batch, frames, n_mels] npy_file = torch.reshape( npy_file, (hp.test.N * hp.test.M // 2, npy_file.size(1), npy_file.size(2))) wav_file_npy = torch.reshape(wav_file_npy, (hp.test.N * hp.test.M // 2, wav_file_npy.size(1), wav_file_npy.size(2))) perm = random.sample(range(0, wav_file_npy.size(0)), wav_file_npy.size(0)) unperm = list(perm) for i, j in enumerate(perm): unperm[j] = i wav_file_npy = wav_file_npy[perm] enrollment_embeddings = embedder_net(npy_file) verification_embeddings = embedder_net(wav_file_npy) verification_embeddings = verification_embeddings[unperm] enrollment_embeddings = torch.reshape( enrollment_embeddings, (hp.test.N, hp.test.M // 2, enrollment_embeddings.size(1))) verification_embeddings = torch.reshape( verification_embeddings, (hp.test.N, hp.test.M // 2, verification_embeddings.size(1))) enrollment_centroids = get_centroids(enrollment_embeddings) sim_matrix = get_cossim(verification_embeddings, enrollment_centroids) # calculating EER diff = 1 EER = 0 EER_thresh = 0 EER_FAR = 0 EER_FRR = 0 for thres in [0.01 * i + 0.5 for i in range(50)]: sim_matrix_thresh = sim_matrix > thres FAR = (sum([ sim_matrix_thresh[i].float().sum() - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (hp.test.N - 1.0) / (float(hp.test.M / 2)) / hp.test.N) FRR = (sum([ hp.test.M / 2 - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (float(hp.test.M / 2)) / hp.test.N) # Save threshold when FAR = FRR (=EER) if diff > abs(FAR - FRR): diff = abs(FAR - FRR) EER = (FAR + FRR) / 2 EER_thresh = thres EER_FAR = FAR EER_FRR = FRR # print("\nEER : %0.2f (thres:%0.2f, FAR:%0.2f, FRR:%0.2f)"%(EER,EER_thresh,EER_FAR,EER_FRR)) sim_matrix_pos = torch.abs(sim_matrix) avg = sim_matrix_pos[0, 0, 1] + sim_matrix_pos[0, 1, 1] + sim_matrix_pos[ 0, 2, 1] + sim_matrix_pos[1, 0, 0] + sim_matrix_pos[ 1, 1, 0] + sim_matrix_pos[1, 2, 0] avg /= 6 # print(sim_matrix) # print(sim_matrix_pos) avg = round(avg.item(), 2) return avg
def test(model_path): utterances_spec = [] for utter_name in os.listdir(predict_folder): print(utter_name) # print(utter_name) if utter_name[-4:] == '.wav': utter_path = os.path.join(predict_folder, utter_name) # path of each utterance utter, sr = librosa.core.load(utter_path, hp.data.sr) # load utterance audio intervals = librosa.effects.split(utter, top_db=30) # voice activity detection utter_min_len = (hp.data.tisv_frame * hp.data.hop + hp.data.window) * hp.data.sr # lower bound of utterance length for interval in intervals: if (interval[1] - interval[0]) > utter_min_len: # If partial utterance is sufficient long, utter_part = utter[interval[0]:interval[1]] # save first and last 180 frames of spectrogram. S = librosa.core.stft(y=utter_part, n_fft=hp.data.nfft, win_length=int(hp.data.window * sr), hop_length=int(hp.data.hop * sr)) S = np.abs(S) ** 2 mel_basis = librosa.filters.mel(sr=hp.data.sr, n_fft=hp.data.nfft, n_mels=hp.data.nmels) S = np.log10(np.dot(mel_basis, S) + 1e-6) # log mel spectrogram of utterances utterances_spec.append(S[:, :hp.data.tisv_frame]) # first 180 frames of partial utterance utterances_spec.append(S[:, -hp.data.tisv_frame:]) # last 180 frames of partial utterance utterances_spec = np.array(utterances_spec) # np.save(os.path.join(hp.data.train_path, "speaker.npy")) test_loader = utterances_spec embedder_net = SpeechEmbedder() embedder_net.load_state_dict(torch.load(model_path)) embedder_net.eval() avg_EER = 0 device = torch.device(hp.device) avg_EER = 0 predict_loader = utterances_spec enrollment_batch, verification_batch = torch.split(predict_loader, int(predict_loader.size(1) / 2), dim=1) enrollment_batch = torch.reshape(enrollment_batch, ( hp.test.N * hp.test.M // 2, enrollment_batch.size(2), enrollment_batch.size(3))) verification_batch = torch.reshape(verification_batch, ( hp.test.N * hp.test.M // 2, verification_batch.size(2), verification_batch.size(3))) perm = random.sample(range(0, verification_batch.size(0)), verification_batch.size(0)) unperm = list(perm) for i, j in enumerate(perm): unperm[j] = i verification_batch = verification_batch[perm] enrollment_embeddings = embedder_net(enrollment_batch) verification_embeddings = embedder_net(verification_batch) verification_embeddings = verification_embeddings[unperm] enrollment_embeddings = torch.reshape(enrollment_embeddings, (hp.test.N, hp.test.M // 2, enrollment_embeddings.size(1))) verification_embeddings = torch.reshape(verification_embeddings, (hp.test.N, hp.test.M // 2, verification_embeddings.size(1))) enrollment_centroids = get_centroids(enrollment_embeddings) sim_matrix = get_cossim(verification_embeddings, enrollment_centroids) diff = 1; EER = 0; EER_thresh = 0; EER_FAR = 0; EER_FRR = 0 for thres in [0.01 * i + 0.5 for i in range(50)]: sim_matrix_thresh = sim_matrix > thres FAR = (sum([sim_matrix_thresh[i].float().sum() - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N))]) / (hp.test.N - 1.0) / (float(hp.test.M / 2)) / hp.test.N) FRR = (sum([hp.test.M / 2 - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N))]) / (float(hp.test.M / 2)) / hp.test.N) if diff > abs(FAR - FRR): diff = abs(FAR - FRR) EER = (FAR + FRR) / 2 EER_thresh = thres EER_FAR = FAR EER_FRR = FRR avg_EER += EER print( "\nEER : %0.2f (thres:%0.2f, FAR:%0.2f, FRR:%0.2f)" % (EER, EER_thresh, EER_FAR, EER_FRR)) print("\n EER across {0} epochs: {:.4f}".format(avg_EER))
def test(model_path): layer_sizes = [hp.model.res_layer for _ in range(hp.model.num_res - 1)] device = torch.device(hp.device) #test_dataset = TestDataset() test_dataset = TrainDataset() test_loader = DataLoader(test_dataset, batch_size=hp.test.N, shuffle=True, num_workers=hp.test.num_workers, drop_last=True) embedder_net = R2Plus1DNet(layer_sizes).to(device) embedder_net.load_state_dict(torch.load(model_path)) embedder_net.eval() avg_EER = 0 for e in range(hp.test.epochs): batch_avg_EER = 0 for batch_id, utters_batch in enumerate(test_loader): assert hp.test.M % 2 == 0 enrollment_batch, verification_batch = torch.split( utters_batch, int(utters_batch.size(1) / 2), dim=1) #enrollment_embeddings = enrollment_batch.to(device) #verification_embeddings = verification_batch.to(device) enrollment_batch = enrollment_batch.to(device) verification_batch = verification_batch.to(device) enrollment_batch = torch.reshape( enrollment_batch, (hp.test.N * hp.test.M // 2, enrollment_batch.size(2), enrollment_batch.size(3), enrollment_batch.size(4), enrollment_batch.size(5))) verification_batch = torch.reshape( verification_batch, (hp.test.N * hp.test.M // 2, verification_batch.size(2), verification_batch.size(3), verification_batch.size(4), verification_batch.size(5))) enrollment_embeddings = embedder_net(enrollment_batch) verification_embeddings = embedder_net(verification_batch) enrollment_embeddings = torch.reshape( enrollment_embeddings, (hp.test.N, hp.test.M // 2, enrollment_embeddings.size(1))) verification_embeddings = torch.reshape( verification_embeddings, (hp.test.N, hp.test.M // 2, verification_embeddings.size(1))) enrollment_centroids = get_centroids(enrollment_embeddings) sim_matrix = get_cossim(verification_embeddings, enrollment_centroids) # calculating EER diff = 1 EER = 0 EER_thresh = 0 EER_FAR = 0 EER_FRR = 0 for thres in [0.01 * i + 0.5 for i in range(50)]: sim_matrix_thresh = sim_matrix > thres FAR = (sum([ sim_matrix_thresh[i].float().sum() - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (hp.test.N - 1.0) / (float(hp.test.M / 2)) / hp.test.N) FRR = (sum([ hp.test.M / 2 - sim_matrix_thresh[i, :, i].float().sum() for i in range(int(hp.test.N)) ]) / (float(hp.test.M / 2)) / hp.test.N) # Save threshold when FAR = FRR (=EER) if diff > abs(FAR - FRR): diff = abs(FAR - FRR) EER = (FAR + FRR) / 2 EER_thresh = thres EER_FAR = FAR EER_FRR = FRR batch_avg_EER += EER print("\nEER : %0.2f (thres:%0.2f, FAR:%0.2f, FRR:%0.2f)" % (EER, EER_thresh, EER_FAR, EER_FRR)) avg_EER += batch_avg_EER / (batch_id + 1) avg_EER = avg_EER / hp.test.epochs print("\n EER across {0} epochs: {1:.4f}".format(hp.test.epochs, avg_EER))