glow.py

# We retain the copyright notice by NVIDIA from the original code. However, we
# we reserve our rights on the modifications based on the original code.
#
# *****************************************************************************
#  Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
#
#  Redistribution and use in source and binary forms, with or without
#  modification, are permitted provided that the following conditions are met:
#      * Redistributions of source code must retain the above copyright
#        notice, this list of conditions and the following disclaimer.
#      * Redistributions in binary form must reproduce the above copyright
#        notice, this list of conditions and the following disclaimer in the
#        documentation and/or other materials provided with the distribution.
#      * Neither the name of the NVIDIA CORPORATION nor the
#        names of its contributors may be used to endorse or promote products
#        derived from this software without specific prior written permission.
#
#  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
#  ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
#  WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
#  DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
#  DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
#  (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
#  LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
#  ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
#  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
#  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# *****************************************************************************
import torch
from torch.autograd import Variable
import torch.nn.functional as F
import numpy as np


@torch.jit.script
def fused_add_tanh_sigmoid_multiply(input_a, input_b, n_channels):
    n_channels_int = n_channels[0]
    in_act = input_a+input_b
    t_act = torch.tanh(in_act[:, :n_channels_int, :])
    s_act = torch.sigmoid(in_act[:, n_channels_int:, :])
    acts = t_act * s_act
    return acts


class Upsample1d(torch.nn.Module):
    def __init__(self, scale=2):
        super(Upsample1d, self).__init__()
        self.scale = scale

    def forward(self, x):
        y = F.interpolate(
            x, scale_factor=self.scale, mode='nearest')
        return y


class SqueezeWaveLoss(torch.nn.Module):
    def __init__(self, sigma=1.0):
        super(SqueezeWaveLoss, self).__init__()
        self.sigma = sigma

    def forward(self, model_output):
        z, log_s_list, log_det_W_list = model_output
        for i, log_s in enumerate(log_s_list):
            if i == 0:
                log_s_total = torch.sum(log_s)
                log_det_W_total = log_det_W_list[i]
            else:
                log_s_total = log_s_total + torch.sum(log_s)
                log_det_W_total += log_det_W_list[i]

        loss = torch.sum(z*z)/(2*self.sigma*self.sigma) - log_s_total - log_det_W_total
        return loss/(z.size(0)*z.size(1)*z.size(2))


class Invertible1x1Conv(torch.nn.Module):
    """
    The layer outputs both the convolution, and the log determinant
    of its weight matrix.  If reverse=True it does convolution with
    inverse
    """
    def __init__(self, c):
        super(Invertible1x1Conv, self).__init__()
        self.conv = torch.nn.Conv1d(c, c, kernel_size=1, stride=1, padding=0,
                                    bias=False)

        # Sample a random orthonormal matrix to initialize weights
        W = torch.qr(torch.FloatTensor(c, c).normal_())[0]

        # Ensure determinant is 1.0 not -1.0
        if torch.det(W) < 0:
            W[:,0] = -1*W[:,0]
        W = W.view(c, c, 1)
        self.conv.weight.data = W

    def forward(self, z, reverse=False):
        # shape
        batch_size, group_size, n_of_groups = z.size()
        W = self.conv.weight.squeeze()

        if reverse:
            if not hasattr(self, 'W_inverse'):
                # Reverse computation
                W_inverse = W.float().inverse()
                W_inverse = Variable(W_inverse[..., None])
                self.W_inverse = W_inverse.half()
            self.W_inverse = self.W_inverse.to(torch.float32)
            z = F.conv1d(z, self.W_inverse, bias=None, stride=1, padding=0)
            return z
        else:
            # Forward computation
            log_det_W = batch_size * n_of_groups * torch.logdet(W)
            z = self.conv(z)
            return z, log_det_W


class WN(torch.nn.Module):
    """
    This is the WaveNet like layer for the affine coupling.  The primary difference
    from WaveNet is the convolutions need not be causal.  There is also no dilation
    size reset.  The dilation only doubles on each layer
    """
    def __init__(self, n_in_channels, n_mel_channels, n_layers, n_channels,
                 kernel_size):
        super(WN, self).__init__()
        assert(kernel_size % 2 == 1)
        assert(n_channels % 2 == 0)
        self.n_layers = n_layers
        self.n_channels = n_channels
        self.in_layers = torch.nn.ModuleList()
        self.res_skip_layers = torch.nn.ModuleList()
        self.upsample = Upsample1d(2)        
        start = torch.nn.Conv1d(n_in_channels, n_channels, 1)
        start = torch.nn.utils.weight_norm(start, name='weight')
        self.start = start
        end = torch.nn.Conv1d(n_channels, 2*n_in_channels, 1)
        end.weight.data.zero_()
        end.bias.data.zero_()
        self.end = end
        
        # cond_layer
        cond_layer = torch.nn.Conv1d(n_mel_channels, 2*n_channels*n_layers, 1)
        self.cond_layer = torch.nn.utils.weight_norm(cond_layer, name='weight')
        for i in range(n_layers):
            dilation = 1
            padding = int((kernel_size*dilation - dilation)/2)
            # depthwise separable convolution
            depthwise = torch.nn.Conv1d(n_channels, n_channels, 3,
                dilation=dilation, padding=padding,
                groups=n_channels)
            pointwise = torch.nn.Conv1d(n_channels, 2*n_channels, 1)
            bn = torch.nn.BatchNorm1d(n_channels) 
            self.in_layers.append(torch.nn.Sequential(bn, depthwise, pointwise))
            # res_skip_layer
            res_skip_layer = torch.nn.Conv1d(n_channels, n_channels, 1)
            res_skip_layer = torch.nn.utils.weight_norm(res_skip_layer, name='weight')
            self.res_skip_layers.append(res_skip_layer)
                        
    def forward(self, forward_input):
        audio, spect = forward_input
        audio = self.start(audio)
        n_channels_tensor = torch.IntTensor([self.n_channels])
        # pass all the mel_spectrograms to cond_layer
        spect = self.cond_layer(spect)
        for i in range(self.n_layers):
            # split the corresponding mel_spectrogram
            spect_offset = i*2*self.n_channels
            spec = spect[:,spect_offset:spect_offset+2*self.n_channels,:]
            if audio.size(2) > spec.size(2):
                cond = self.upsample(spec)
            else:
                cond = spec
            acts = fused_add_tanh_sigmoid_multiply(
                self.in_layers[i](audio),
                cond, 
                n_channels_tensor)
            # res_skip
            res_skip_acts = self.res_skip_layers[i](acts)
            audio = audio + res_skip_acts
        return self.end(audio)


class SqueezeWave(torch.nn.Module):
    def __init__(self, n_mel_channels, n_flows, n_audio_channel, n_early_every,
                 n_early_size, WN_config):
        super(SqueezeWave, self).__init__()        
        assert(n_audio_channel % 2 == 0)
        self.n_flows = n_flows
        self.n_audio_channel = n_audio_channel
        self.n_early_every = n_early_every
        self.n_early_size = n_early_size
        self.WN = torch.nn.ModuleList()
        self.convinv = torch.nn.ModuleList()

        n_half = int(n_audio_channel / 2)
        # Set up layers with the right sizes based on how many dimensions
        # have been output already
        n_remaining_channels = n_audio_channel
        for k in range(n_flows):
            if k % self.n_early_every == 0 and k > 0:
                n_half = n_half - int(self.n_early_size/2)
                n_remaining_channels = n_remaining_channels - self.n_early_size
            self.convinv.append(Invertible1x1Conv(n_remaining_channels))
            self.WN.append(WN(n_half, n_mel_channels, **WN_config))

        self.n_remaining_channels = n_remaining_channels  # Useful during inference

    def forward(self, forward_input):
        """
        forward_input[0] = mel_spectrogram:  batch x n_mel_channels x frames
        forward_input[1] = audio: batch x time
        """
        spect, audio = forward_input

        audio = audio.unfold(
            1, self.n_audio_channel, self.n_audio_channel).permute(0, 2, 1)
        output_audio = []
        log_s_list = []
        log_det_W_list = []

        for k in range(self.n_flows):
            if k % self.n_early_every == 0 and k > 0:
                output_audio.append(audio[:,:self.n_early_size,:])
                audio = audio[:,self.n_early_size:,:]

            audio, log_det_W = self.convinv[k](audio)
            log_det_W_list.append(log_det_W)

            n_half = int(audio.size(1)/2)
            audio_0 = audio[:,:n_half,:]
            audio_1 = audio[:,n_half:,:]

            output = self.WN[k]((audio_0, spect))
            log_s = output[:, n_half:, :]
            b = output[:, :n_half, :]

            audio_1 = (torch.exp(log_s))*audio_1 + b
            log_s_list.append(log_s)
            audio = torch.cat([audio_0, audio_1], 1)

        output_audio.append(audio)
        return torch.cat(output_audio, 1), log_s_list, log_det_W_list

    def infer(self, spect, sigma=1.0):
        spect_size = spect.size()
        l = spect.size(2)*(256 // self.n_audio_channel)

        spect = spect.to(torch.float32)
        if spect.type() == 'torch.HalfTensor':
            audio = torch.HalfTensor(spect.size(0),
                                          self.n_remaining_channels,
                                          l).normal_()
        else:
            audio = torch.FloatTensor(spect.size(0),
                                           self.n_remaining_channels,
                                           l).normal_()

        for k in reversed(range(self.n_flows)):
            n_half = int(audio.size(1)/2)
            audio_0 = audio[:,:n_half,:]
            audio_1 = audio[:,n_half:,:]
            output = self.WN[k]((audio_0, spect))

            s = output[:, n_half:, :]
            b = output[:, :n_half, :]
            audio_1 = (audio_1 - b)/torch.exp(s)
            audio = torch.cat([audio_0, audio_1],1)

            audio = self.convinv[k](audio, reverse=True)

            if k % self.n_early_every == 0 and k > 0:
                if spect.type() == 'torch.HalfTensor':
                    z = torch.HalfTensor(spect.size(0), self.n_early_size, l).normal_()
                else:
                    z = torch.FloatTensor(spect.size(0), self.n_early_size, l).normal_()
                audio = torch.cat((sigma*z, audio),1)

        audio = audio.permute(0,2,1).contiguous().view(audio.size(0), -1).data
        return audio


    @staticmethod
    def remove_weightnorm(model):
        squeezewave = model
        for WN in squeezewave.WN:
            WN.start = torch.nn.utils.remove_weight_norm(WN.start)
            WN.in_layers = remove_batch_norm(WN.in_layers)
            WN.cond_layer = torch.nn.utils.remove_weight_norm(WN.cond_layer)
            WN.res_skip_layers = remove(WN.res_skip_layers) 
        return squeezewave

def fuse_conv_and_bn(conv, bn):
    fusedconv = torch.nn.Conv1d(
            conv.in_channels,
            conv.out_channels,
            kernel_size = conv.kernel_size,
            padding=conv.padding,
            bias=True,
            groups=conv.groups)
    w_conv = conv.weight.clone().view(conv.out_channels, -1)
    w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps+bn.running_var)))
    w_bn = w_bn.clone()
    fusedconv.weight.data = torch.mm(w_bn, w_conv).view(fusedconv.weight.size())
    if conv.bias is not None:
        b_conv = conv.bias
    else:
        b_conv = torch.zeros( conv.weight.size(0) )
    b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps))
    b_bn = torch.unsqueeze(b_bn, 1)
    bn_3 = b_bn.expand(-1, 3)
    b = torch.matmul(w_conv, torch.transpose(bn_3, 0, 1))[range(b_bn.size()[0]), range(b_bn.size()[0])]
    fusedconv.bias.data = ( b_conv + b )
    return fusedconv

def remove_batch_norm(conv_list):
    new_conv_list = torch.nn.ModuleList()
    for old_conv in conv_list:
        depthwise = fuse_conv_and_bn(old_conv[1], old_conv[0])
        pointwise = old_conv[2]
        new_conv_list.append(torch.nn.Sequential(depthwise, pointwise))
    return new_conv_list

def remove(conv_list):
    new_conv_list = torch.nn.ModuleList()
    for old_conv in conv_list:
        old_conv = torch.nn.utils.remove_weight_norm(old_conv)
        new_conv_list.append(old_conv)
    return new_conv_list