inference_batch_from_csv.py

###############################################################################
#
#  Copyright (c) 2020, NVIDIA CORPORATION. All rights reserved.
#  Licensed under the Apache License, Version 2.0 (the "License");
#  you may not use this file except in compliance with the License.
#  You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
#  Unless required by applicable law or agreed to in writing, software
#  distributed under the License is distributed on an "AS IS" BASIS,
#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#  See the License for the specific language governing permissions and
#  limitations under the License.
#
###############################################################################
import matplotlib
matplotlib.use("Agg")
import matplotlib.pylab as plt

import os
import argparse
import json
import sys
import numpy as np
import torch
import csv


from flowtron import Flowtron
from torch.utils.data import DataLoader
from data import Data
from train import update_params

sys.path.insert(0, "tacotron2")
sys.path.insert(0, "tacotron2/waveglow")
from glow import WaveGlow
from scipy.io.wavfile import write


def load_text_batch(csv_path, text_lines, text_ids):
    with open(csv_path) as csv_file:
        csv_reader = csv.reader(csv_file, delimiter='|')
        line_count = 0
        for row in csv_reader:
            line_count += 1
            if line_count == 1:
                continue
            else:
                text_ids.append(row[0])
                text_lines.append(row[1])
    print("Batch: loaded {} lines from file.".format(line_count))

def infer(flowtron_path, waveglow_path, output_dir, csv_path, speaker_id, n_frames,
          sigma, gate_threshold, seed):
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)

    # load waveglow
    waveglow = torch.load(waveglow_path)['model'].cuda().eval()
    waveglow.cuda().half()
    for k in waveglow.convinv:
        k.float()
    waveglow.eval()

    # load flowtron
    try:
        model = Flowtron(**model_config).cuda()
        state_dict = torch.load(flowtron_path, map_location='cpu')['state_dict']
        model.load_state_dict(state_dict)
    except KeyError:
        # model saved by train.py module
        # do not need to load state dict
        # and can be used directly
        model = torch.load(flowtron_path)['model']
    model.eval()
    print("Loaded checkpoint '{}')" .format(flowtron_path))

    ignore_keys = ['training_files', 'validation_files']
    trainset = Data(
        data_config['training_files'],
        **dict((k, v) for k, v in data_config.items() if k not in ignore_keys))

    speaker_vecs = trainset.get_speaker_id(speaker_id).cuda()
    speaker_vecs = speaker_vecs[None]

    with torch.no_grad():
        residual = torch.cuda.FloatTensor(1, 80, n_frames).normal_() * sigma

    text_lines = []
    text_ids = []
    load_text_batch(csv_path, text_lines, text_ids)

    for idx in range(len(text_lines)):
        text = text_lines[idx]
        file_suffix = text_ids[idx] # + (text[:20] + "..") if len(text) > 20 else text
        text = trainset.get_text(text).cuda()

        text = text[None]

        with torch.no_grad():
            #residual = torch.cuda.FloatTensor(1, 80, n_frames).normal_() * sigma
            mels, attentions = model.infer(
                residual, speaker_vecs, text, gate_threshold=gate_threshold)

        for k in range(len(attentions)):
            attention = torch.cat(attentions[k]).cpu().numpy()
            fig, axes = plt.subplots(1, 2, figsize=(16, 4))
            #axes[0].imshow(mels[0].cpu().numpy(), origin='bottom', aspect='auto')
            #axes[1].imshow(attention[:, 0].transpose(), origin='bottom', aspect='auto')
            axes[0].imshow(mels[0].cpu().numpy(), origin='lower', aspect='auto')
            axes[1].imshow(attention[:, 0].transpose(), origin='lower', aspect='auto')
            #fig.savefig(os.path.join(output_dir, 'sid{}_sigma{}_frames{}_attnlayer{}.png'.format(speaker_id, sigma, n_frames, k)))
            plt.close("all")

        with torch.no_grad():
            audio = waveglow.infer(mels.half(), sigma=0.8).float()

        audio = audio.cpu().numpy()[0]
        # normalize audio for now
        audio = audio / np.abs(audio).max()

        write(os.path.join(output_dir, 'sid{}_s{}_{}.wav'.format(speaker_id, sigma, file_suffix)),
              data_config['sampling_rate'], audio)

        print("Ready! audio: {}, text: {}".format(audio.shape, text_lines[idx]))
    print("Finished inferencing {} lines.".format(len(text_lines)))


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument('-c', '--config', type=str,
                        help='JSON file for configuration')
    parser.add_argument('-p', '--params', nargs='+', default=[])
    parser.add_argument('-f', '--flowtron_path',
                        help='Path to flowtron state dict', type=str)
    parser.add_argument('-w', '--waveglow_path',
                        help='Path to waveglow state dict', type=str)
    parser.add_argument("--csv_path", default='batch_lines.csv', type=str, help='Path to csv file')
    parser.add_argument('-i', '--id', help='Speaker id', type=int)
    parser.add_argument('-n', '--n_frames', help='Number of frames',
                        default=400, type=int)
    parser.add_argument('-o', "--output_dir", default="results/")
    parser.add_argument("-s", "--sigma", default=0.5, type=float)
    parser.add_argument("-g", "--gate", default=0.5, type=float)
    parser.add_argument("--seed", default=1234, type=int)
    args = parser.parse_args()

    # Parse configs.  Globals nicer in this case
    with open(args.config) as f:
        data = f.read()

    global config
    config = json.loads(data)
    update_params(config, args.params)

    data_config = config["data_config"]
    global model_config
    model_config = config["model_config"]

    # Make directory if it doesn't exist
    if not os.path.isdir(args.output_dir):
        os.makedirs(args.output_dir)
        os.chmod(args.output_dir, 0o775)

    torch.backends.cudnn.enabled = True
    torch.backends.cudnn.benchmark = False
    infer(args.flowtron_path, args.waveglow_path, args.output_dir, args.csv_path,
          args.id, args.n_frames, args.sigma, args.gate, args.seed)