realtime-voice-conversion based on last year's work

This is based on the work I did in (July - October)ish 2019 to get real time voice conversion working. I've removed it from the `voder` library and made it into an isolable program. I'll work on extending the capabilities.
2026-10-22 07:55:47 +00:00 · 2020-09-11 09:06:48 -04:00
commit 4a222ad815
21 changed files with 3102 additions and 0 deletions
@@ -0,0 +1,238 @@
+import librosa
+import numpy as np
+import os
+import pyworld
+
+def load_wavs(wav_dir, sr):
+
+    wavs = list()
+    for file in os.listdir(wav_dir):
+        file_path = os.path.join(wav_dir, file)
+        wav, _ = librosa.load(file_path, sr = sr, mono = True)
+        #wav = wav.astype(np.float64)
+        wavs.append(wav)
+
+    return wavs
+
+def world_decompose(wav, fs, frame_period = 5.0):
+
+    # Decompose speech signal into f0, spectral envelope and aperiodicity using WORLD
+    wav = wav.astype(np.float64)
+    f0, timeaxis = pyworld.harvest(wav, fs, frame_period = frame_period, f0_floor = 71.0, f0_ceil = 800.0)
+    sp = pyworld.cheaptrick(wav, f0, timeaxis, fs)
+    ap = pyworld.d4c(wav, f0, timeaxis, fs)
+
+    return f0, timeaxis, sp, ap
+
+def world_encode_spectral_envelop(sp, fs, dim = 24):
+
+    # Get Mel-cepstral coefficients (MCEPs)
+
+    #sp = sp.astype(np.float64)
+    coded_sp = pyworld.code_spectral_envelope(sp, fs, dim)
+
+    return coded_sp
+
+def world_decode_spectral_envelop(coded_sp, fs):
+
+    fftlen = pyworld.get_cheaptrick_fft_size(fs)
+    #coded_sp = coded_sp.astype(np.float32)
+    #coded_sp = np.ascontiguousarray(coded_sp)
+    decoded_sp = pyworld.decode_spectral_envelope(coded_sp, fs, fftlen)
+
+    return decoded_sp
+
+
+def world_encode_data(wavs, fs, frame_period = 5.0, coded_dim = 24):
+
+    f0s = list()
+    timeaxes = list()
+    sps = list()
+    aps = list()
+    coded_sps = list()
+
+    for wav in wavs:
+        f0, timeaxis, sp, ap = world_decompose(wav = wav, fs = fs, frame_period = frame_period)
+        coded_sp = world_encode_spectral_envelop(sp = sp, fs = fs, dim = coded_dim)
+        f0s.append(f0)
+        timeaxes.append(timeaxis)
+        sps.append(sp)
+        aps.append(ap)
+        coded_sps.append(coded_sp)
+
+    return f0s, timeaxes, sps, aps, coded_sps
+
+
+def transpose_in_list(lst):
+
+    transposed_lst = list()
+    for array in lst:
+        transposed_lst.append(array.T)
+    return transposed_lst
+
+
+def world_decode_data(coded_sps, fs):
+
+    decoded_sps =  list()
+
+    for coded_sp in coded_sps:
+        decoded_sp = world_decode_spectral_envelop(coded_sp, fs)
+        decoded_sps.append(decoded_sp)
+
+    return decoded_sps
+
+
+def world_speech_synthesis(f0, decoded_sp, ap, fs, frame_period):
+
+    #decoded_sp = decoded_sp.astype(np.float64)
+    wav = pyworld.synthesize(f0, decoded_sp, ap, fs, frame_period)
+    # Librosa could not save wav if not doing so
+    wav = wav.astype(np.float32)
+
+    return wav
+
+
+def world_synthesis_data(f0s, decoded_sps, aps, fs, frame_period):
+
+    wavs = list()
+
+    for f0, decoded_sp, ap in zip(f0s, decoded_sps, aps):
+        wav = world_speech_synthesis(f0, decoded_sp, ap, fs, frame_period)
+        wavs.append(wav)
+
+    return wavs
+
+
+def coded_sps_normalization_fit_transoform(coded_sps):
+
+    coded_sps_concatenated = np.concatenate(coded_sps, axis = 1)
+    coded_sps_mean = np.mean(coded_sps_concatenated, axis = 1, keepdims = True)
+    coded_sps_std = np.std(coded_sps_concatenated, axis = 1, keepdims = True)
+
+    coded_sps_normalized = list()
+    for coded_sp in coded_sps:
+        coded_sps_normalized.append((coded_sp - coded_sps_mean) / coded_sps_std)
+    
+    return coded_sps_normalized, coded_sps_mean, coded_sps_std
+
+def coded_sps_normalization_transoform(coded_sps, coded_sps_mean, coded_sps_std):
+
+    coded_sps_normalized = list()
+    for coded_sp in coded_sps:
+        coded_sps_normalized.append((coded_sp - coded_sps_mean) / coded_sps_std)
+    
+    return coded_sps_normalized
+
+def coded_sps_normalization_inverse_transoform(normalized_coded_sps, coded_sps_mean, coded_sps_std):
+
+    coded_sps = list()
+    for normalized_coded_sp in normalized_coded_sps:
+        coded_sps.append(normalized_coded_sp * coded_sps_std + coded_sps_mean)
+
+    return coded_sps
+
+def coded_sp_padding(coded_sp, multiple = 4):
+
+    num_features = coded_sp.shape[0]
+    num_frames = coded_sp.shape[1]
+    num_frames_padded = int(np.ceil(num_frames / multiple)) * multiple
+    num_frames_diff = num_frames_padded - num_frames
+    num_pad_left = num_frames_diff // 2
+    num_pad_right = num_frames_diff - num_pad_left
+    coded_sp_padded = np.pad(coded_sp, ((0, 0), (num_pad_left, num_pad_right)), 'constant', constant_values = 0)
+
+    return coded_sp_padded
+
+def wav_padding(wav, sr, frame_period, multiple = 4):
+
+    assert wav.ndim == 1 
+    num_frames = len(wav)
+    num_frames_padded = int((np.ceil((np.floor(num_frames / (sr * frame_period / 1000)) + 1) / multiple + 1) * multiple - 1) * (sr * frame_period / 1000))
+    num_frames_diff = num_frames_padded - num_frames
+    num_pad_left = num_frames_diff // 2
+    num_pad_right = num_frames_diff - num_pad_left
+    wav_padded = np.pad(wav, (num_pad_left, num_pad_right), 'constant', constant_values = 0)
+
+    return wav_padded
+
+
+def logf0_statistics(f0s):
+
+    log_f0s_concatenated = np.ma.log(np.concatenate(f0s))
+    log_f0s_mean = log_f0s_concatenated.mean()
+    log_f0s_std = log_f0s_concatenated.std()
+
+    return log_f0s_mean, log_f0s_std
+
+def pitch_conversion(f0, mean_log_src, std_log_src, mean_log_target, std_log_target):
+
+    # Logarithm Gaussian normalization for Pitch Conversions
+    f0_converted = np.exp((np.log(f0) - mean_log_src) / std_log_src * std_log_target + mean_log_target)
+
+    return f0_converted
+
+def wavs_to_specs(wavs, n_fft = 1024, hop_length = None):
+
+    stfts = list()
+    for wav in wavs:
+        stft = librosa.stft(wav, n_fft = n_fft, hop_length = hop_length)
+        stfts.append(stft)
+
+    return stfts
+
+
+def wavs_to_mfccs(wavs, sr, n_fft = 1024, hop_length = None, n_mels = 128, n_mfcc = 24):
+
+    mfccs = list()
+    for wav in wavs:
+        mfcc = librosa.feature.mfcc(y = wav, sr = sr, n_fft = n_fft, hop_length = hop_length, n_mels = n_mels, n_mfcc = n_mfcc)
+        mfccs.append(mfcc)
+
+    return mfccs
+
+
+def mfccs_normalization(mfccs):
+
+    mfccs_concatenated = np.concatenate(mfccs, axis = 1)
+    mfccs_mean = np.mean(mfccs_concatenated, axis = 1, keepdims = True)
+    mfccs_std = np.std(mfccs_concatenated, axis = 1, keepdims = True)
+
+    mfccs_normalized = list()
+    for mfcc in mfccs:
+        mfccs_normalized.append((mfcc - mfccs_mean) / mfccs_std)
+    
+    return mfccs_normalized, mfccs_mean, mfccs_std
+
+
+def sample_train_data(dataset_A, dataset_B, n_frames = 128):
+
+    num_samples = min(len(dataset_A), len(dataset_B))
+    train_data_A_idx = np.arange(len(dataset_A))
+    train_data_B_idx = np.arange(len(dataset_B))
+    np.random.shuffle(train_data_A_idx)
+    np.random.shuffle(train_data_B_idx)
+    train_data_A_idx_subset = train_data_A_idx[:num_samples]
+    train_data_B_idx_subset = train_data_B_idx[:num_samples]
+
+    train_data_A = list()
+    train_data_B = list()
+
+    for idx_A, idx_B in zip(train_data_A_idx_subset, train_data_B_idx_subset):
+        data_A = dataset_A[idx_A]
+        frames_A_total = data_A.shape[1]
+        assert frames_A_total >= n_frames
+        start_A = np.random.randint(frames_A_total - n_frames + 1)
+        end_A = start_A + n_frames
+        train_data_A.append(data_A[:,start_A:end_A])
+
+        data_B = dataset_B[idx_B]
+        frames_B_total = data_B.shape[1]
+        assert frames_B_total >= n_frames
+        start_B = np.random.randint(frames_B_total - n_frames + 1)
+        end_B = start_B + n_frames
+        train_data_B.append(data_B[:,start_B:end_B])
+
+    train_data_A = np.array(train_data_A)
+    train_data_B = np.array(train_data_B)
+
+    return train_data_A, train_data_B