Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

linter fixes

This commit is contained in:
erogol 2020-06-19 14:46:13 +02:00
parent 6b2ff08239
commit ce076330bf
8 changed files with 46 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vocoder/tests/test_melgan_discriminator.py
View File

@ -20,7 +20,7 @@ def test_melgan_multi_scale_discriminator():
scores, feats = model(dummy_input)
assert len(scores) == 3
assert len(scores) == len(feats)
assert np.all(scores[0].shape == (4, 1, 16))
assert np.all(scores[0].shape == (4, 1, 64))
assert np.all(feats[0][0].shape == (4, 16, 4096))
assert np.all(feats[0][1].shape == (4, 64, 1024))
assert np.all(feats[0][2].shape == (4, 256, 256))

7
vocoder/tf/convert_melgan_torch_to_tf.py
View File

@ -1,7 +1,5 @@
import argparse
import os
import sys
from pprint import pprint
import numpy as np
import tensorflow as tf
@ -9,8 +7,6 @@ import torch
from fuzzywuzzy import fuzz
from TTS.utils.io import load_config
from TTS.vocoder.tf.models.multiband_melgan_generator import \
MultibandMelganGenerator
from TTS.vocoder.tf.utils.convert_torch_to_tf_utils import (
compare_torch_tf, convert_tf_name, transfer_weights_torch_to_tf)
from TTS.vocoder.tf.utils.generic_utils import \
@ -55,7 +51,7 @@ model_tf = setup_tf_generator(c)
common_sufix = '/.ATTRIBUTES/VARIABLE_VALUE'
# get tf_model graph by passing an input
# B x D x T
dummy_input = tf.random.uniform((7, 80, 64))
dummy_input = tf.random.uniform((7, 80, 64), dtype=tf.float32)
mel_pred = model_tf(dummy_input, training=False)
# get tf variables
@ -64,6 +60,7 @@ tf_vars = model_tf.weights
# match variable names with fuzzy logic
torch_var_names = list(state_dict.keys())
tf_var_names = [we.name for we in model_tf.weights]
var_map = []
for tf_name in tf_var_names:
# skip re-mapped layer names
if tf_name in [name[0] for name in var_map]:

24
vocoder/tf/layers/melgan.py
View File

@ -28,18 +28,24 @@ class ResidualStack(tf.keras.layers.Layer):
tf.keras.layers.LeakyReLU(0.2),
ReflectionPad1d(layer_padding),
tf.keras.layers.Conv2D(filters=channels,
kernel_size=(kernel_size, 1),
dilation_rate=(layer_dilation, 1),
use_bias=True,
padding='valid',
name=f'blocks.{idx}.{num_layers}'),
kernel_size=(kernel_size, 1),
dilation_rate=(layer_dilation, 1),
use_bias=True,
padding='valid',
name=f'blocks.{idx}.{num_layers}'),
tf.keras.layers.LeakyReLU(0.2),
tf.keras.layers.Conv2D(filters=channels, kernel_size=(1, 1), use_bias=True, name=f'blocks.{idx}.{num_layers + 2}')
tf.keras.layers.Conv2D(filters=channels,
kernel_size=(1, 1),
use_bias=True,
name=f'blocks.{idx}.{num_layers + 2}')
]
self.blocks.append(block)
self.shortcuts = [
tf.keras.layers.Conv2D(channels, kernel_size=1, use_bias=True, name=f'shortcuts.{i}')
for i in range(num_res_blocks)
tf.keras.layers.Conv2D(channels,
kernel_size=1,
use_bias=True,
name=f'shortcuts.{i}')
for i in range(num_res_blocks)
]
def call(self, x):
@ -47,6 +53,6 @@ class ResidualStack(tf.keras.layers.Layer):
for block, shortcut in zip(self.blocks, self.shortcuts):
res = shortcut(x)
for layer in block:
x = layer(x)
x = layer(x)
x += res
return x

6
vocoder/tf/layers/pqmf.py
View File

@ -40,7 +40,7 @@ class PQMF(tf.keras.layers.Layer):
x : B x 1 x T
"""
x = tf.transpose(x, perm=[0, 2, 1])
x = tf.pad(x, [[0, 0], [self.taps // 2, self.taps // 2], [0, 0]])
x = tf.pad(x, [[0, 0], [self.taps // 2, self.taps // 2], [0, 0]], constant_values=0.0)
x = tf.nn.conv1d(x, self.H, stride=1, padding='VALID')
x = tf.nn.conv1d(x,
self.updown_filter,
@ -60,7 +60,7 @@ class PQMF(tf.keras.layers.Layer):
strides=self.N,
output_shape=(tf.shape(x)[0], tf.shape(x)[1] * self.N,
self.N))
x = tf.pad(x, [[0, 0], [self.taps // 2, self.taps // 2], [0, 0]])
x = tf.pad(x, [[0, 0], [self.taps // 2, self.taps // 2], [0, 0]], constant_values=0.0)
x = tf.nn.conv1d(x, self.G, stride=1, padding="VALID")
x = tf.transpose(x, perm=[0, 2, 1])
return x
return x

49
vocoder/tf/models/melgan_generator.py
View File

@ -8,7 +8,7 @@ import tensorflow as tf
from TTS.vocoder.tf.layers.melgan import ResidualStack, ReflectionPad1d
class MelganGenerator(tf.keras.models.Model):
class MelganGenerator(tf.keras.models.Model): # pylint: disable=too-many-ancestors
""" Melgan Generator TF implementation dedicated for inference with no
weight norm """
def __init__(self,
@ -21,6 +21,8 @@ class MelganGenerator(tf.keras.models.Model):
num_res_blocks=3):
super(MelganGenerator, self).__init__()
self.in_channels = in_channels
# assert model parameters
assert (proj_kernel -
1) % 2 == 0, " [!] proj_kernel should be an odd number."
@ -34,11 +36,11 @@ class MelganGenerator(tf.keras.models.Model):
self.initial_layer = [
ReflectionPad1d(base_padding),
tf.keras.layers.Conv2D(filters=base_channels,
kernel_size=(proj_kernel, 1),
strides=1,
padding='valid',
use_bias=True,
name="1")
kernel_size=(proj_kernel, 1),
strides=1,
padding='valid',
use_bias=True,
name="1")
]
num_layers = 3 # count number of layers for layer naming
@ -48,23 +50,21 @@ class MelganGenerator(tf.keras.models.Model):
layer_out_channels = base_channels // (2**(idx + 1))
layer_filter_size = upsample_factor * 2
layer_stride = upsample_factor
layer_output_padding = upsample_factor % 2
# layer_output_padding = upsample_factor % 2
self.upsample_layers += [
tf.keras.layers.LeakyReLU(act_slope),
tf.keras.layers.Conv2DTranspose(
filters=layer_out_channels,
kernel_size=(layer_filter_size, 1),
strides=(layer_stride, 1),
padding='same',
# output_padding=layer_output_padding,
use_bias=True,
name=f'{num_layers}'),
ResidualStack(
channels=layer_out_channels,
num_res_blocks=num_res_blocks,
kernel_size=res_kernel,
name=f'layers.{num_layers + 1}'
)
filters=layer_out_channels,
kernel_size=(layer_filter_size, 1),
strides=(layer_stride, 1),
padding='same',
# output_padding=layer_output_padding,
use_bias=True,
name=f'{num_layers}'),
ResidualStack(channels=layer_out_channels,
num_res_blocks=num_res_blocks,
kernel_size=res_kernel,
name=f'layers.{num_layers + 1}')
]
num_layers += num_res_blocks - 1
@ -74,15 +74,12 @@ class MelganGenerator(tf.keras.models.Model):
self.final_layers = [
ReflectionPad1d(base_padding),
tf.keras.layers.Conv2D(filters=out_channels,
kernel_size=(proj_kernel, 1),
use_bias=True,
name=f'layers.{num_layers + 1}'),
kernel_size=(proj_kernel, 1),
use_bias=True,
name=f'layers.{num_layers + 1}'),
tf.keras.layers.Activation("tanh")
]
# self.initial_layer = tf.keras.models.Sequential(self.initial_layer)
# self.upsample_layers = tf.keras.models.Sequential(self.upsample_layers)
# self.final_layers = tf.keras.models.Sequential(self.final_layers)
# self.model_layers = tf.keras.models.Sequential(self.initial_layer + self.upsample_layers + self.final_layers, name="layers")
self.model_layers = self.initial_layer + self.upsample_layers + self.final_layers

2
vocoder/tf/models/multiband_melgan_generator.py
View File

@ -4,7 +4,7 @@ from TTS.vocoder.tf.models.melgan_generator import MelganGenerator
from TTS.vocoder.tf.layers.pqmf import PQMF
class MultibandMelganGenerator(MelganGenerator):
class MultibandMelganGenerator(MelganGenerator): # pylint: disable=too-many-ancestors
def __init__(self,
in_channels=80,
out_channels=4,

5
vocoder/tf/utils/convert_torch_to_tf_utils.py
View File

@ -30,10 +30,7 @@ def transfer_weights_torch_to_tf(tf_vars, var_map_dict, state_dict):
# if tuple, it is a bias variable
if 'kernel' in tf_var.name:
torch_weight = state_dict[torch_var_name]
try:
numpy_weight = torch_weight.permute([2, 1, 0]).numpy()[:, None, :, :]
except:
breakpoint()
numpy_weight = torch_weight.permute([2, 1, 0]).numpy()[:, None, :, :]
if 'bias' in tf_var.name:
torch_weight = state_dict[torch_var_name]
numpy_weight = torch_weight

2
vocoder/tf/utils/generic_utils.py
View File

@ -1,7 +1,5 @@
import re
import importlib
import numpy as np
from matplotlib import pyplot as plt
def to_camel(text):