新增模型vits-svc模型

paddlemix/models/vits-svc/configs/base.yaml

@@ -0,0 +1,72 @@
+train:
+ model: "sovits"
+ seed: 1234
+ epochs: 10000
+ learning_rate: 5e-5
+ betas: [0.8, 0.99]
+ lr_decay: 0.999875
+ eps: 1e-9
+ batch_size: 8
+ accum_step: 2
+ c_stft: 9
+ c_mel: 1.
+ c_kl: 0.2
+ port: 8001
+ pretrain: "./vits_pretrain/sovits5.0.pretrain.pth"
+#############################
+data: 
+ training_files: "files/train.txt"
+ validation_files: "files/valid.txt"
+ segment_size: 8000 # WARNING: base on hop_length
+ max_wav_value: 32768.0
+ sampling_rate: 32000
+ filter_length: 1024
+ hop_length: 320
+ win_length: 1024
+ mel_channels: 100
+ mel_fmin: 50.0
+ mel_fmax: 16000.0
+#############################
+vits:
+ ppg_dim: 1280
+ vec_dim: 256
+ spk_dim: 256
+ gin_channels: 256
+ inter_channels: 192
+ hidden_channels: 192
+ filter_channels: 640
+#############################
+gen:
+ upsample_input: 192
+ upsample_rates: [5,4,4,2,2]
+ upsample_kernel_sizes: [15,8,8,4,4]
+ upsample_initial_channel: 320
+ resblock_kernel_sizes: [3,7,11]
+ resblock_dilation_sizes: [[1,3,5], [1,3,5], [1,3,5]]
+#############################
+mpd:
+ periods: [2,3,5,7,11]
+ kernel_size: 5
+ stride: 3
+ use_spectral_norm: False
+ lReLU_slope: 0.2
+#############################
+mrd:
+ resolutions: "[(1024, 120, 600), (2048, 240, 1200), (4096, 480, 2400), (512, 50, 240)]" # (filter_length, hop_length, win_length)
+ use_spectral_norm: False
+ lReLU_slope: 0.2
+#############################
+log:
+ info_interval: 100
+ eval_interval: 1
+ save_interval: 5
+ num_audio: 6
+ pth_dir: 'chkpt'
+ log_dir: 'logs'
+ keep_ckpts: 0
+#############################
+dist_config:
+ dist_backend: "nccl"
+ dist_url: "tcp://localhost:54321"
+ world_size: 1
+

paddlemix/models/vits-svc/crepe/__init__.py

@@ -0,0 +1,8 @@
+from . import decode
+from .core import *
+from .model import Crepe
+from . import convert
+from . import filter
+from . import load
+# from . import loudness
+# from . import threshold

paddlemix/models/vits-svc/crepe/convert.py

@@ -0,0 +1,58 @@
+import scipy
+import paddle
+import math
+import crepe
+
+
+###############################################################################
+# Pitch unit conversions
+###############################################################################
+
+
+def bins_to_cents(bins):
+ """Converts pitch bins to cents"""
+ cents = crepe.CENTS_PER_BIN * bins + 1997.3794084376191
+
+ # Trade quantization error for noise
+ return dither(cents)
+
+
+def bins_to_frequency(bins):
+ """Converts pitch bins to frequency in Hz"""
+ return cents_to_frequency(bins_to_cents(bins))
+
+
+def cents_to_bins(cents, quantize_fn=math.floor):
+ """Converts cents to pitch bins"""
+ bins = (cents - 1997.3794084376191) / crepe.CENTS_PER_BIN
+ return quantize_fn(bins)
+
+
+def cents_to_frequency(cents):
+ """Converts cents to frequency in Hz"""
+ return 10 * 2 ** (cents / 1200)
+
+
+def frequency_to_bins(frequency, quantize_fn=math.floor):
+ """Convert frequency in Hz to pitch bins"""
+ return cents_to_bins(frequency_to_cents(frequency), quantize_fn)
+
+
+def frequency_to_cents(frequency):
+ """Convert frequency in Hz to cents"""
+ return 1200 * math.log2(frequency / 10.)
+
+
+# ###############################################################################
+# # Utilities
+# ###############################################################################
+
+
+def dither(cents):
+ """Dither the predicted pitch in cents to remove quantization error"""
+ noise = scipy.stats.triang.rvs(c=0.5,
+ loc=-crepe.CENTS_PER_BIN,
+ scale=2 * crepe.CENTS_PER_BIN,
+ size=cents.shape)
+ # return cents + cents.new_tensor(noise)
+ return cents + paddle.to_tensor(noise, dtype=cents.dtype, stop_gradient=cents.stop_gradient)