Speaker Identification

1.Goal

根据给定的语音内容，识别出说话者是谁

2.Data formats

2.1data directory

目录下有三个json文件和很多pt文件，三个json文件作用标注在下图中，pt文件就是语音内容。

mapping文件

metadata文件
n_mels:The demission of mel-spectrogram（特征数是40）
speakers: A dictionary

• key: speaker id
• value: feature_path and mel_len
  可以发现，pt文件内容长度不一样，所以后期需要我们自己统一长度
  

testdata文件

3.DataSet

构建自己的Dataset类，需要知道数据集的地址，由于每个数据长度不一样，所以还要规定数据的长度，返回值应该是，语音数据和对应的speaker的label

Dataset有三个重要的方法需要重写

• __init__(): 初始化，一般用于读取给定的数据到内存，在该任务中，需要从给定的路径中，读取语音path和label到data数组，读取mapping将speaker的id映射为对应序号。
• __getitem__(): 该方法根据传入的参数返回指定id的数据，在该任务中，对于传入的id，我们从data数组取出path和label，并将指定path的语音数据读出后分割，最后返回语音片段和对应的label。
• __len__(): 该方法返回数据集的长度，在该任务中，数据集的长度即data数组的长度。

代码如下：

import os
import json
import torch
import random
from pathlib import Path
from torch.nn.utils.rnn import pad_sequence
from torch.utils.data import Dataset


class myDataset(Dataset):
    def __init__(self, data_dir, segment_len=128):
        self.data_dir = data_dir
        self.segment_len = segment_len

        # Load the mapping from speaker neme to their corresponding id.
        mapping_path = Path(data_dir) / "mapping.json"
        mapping = json.load(mapping_path.open())
        self.speaker2id = mapping["speaker2id"] # 在mapping.json 文件中 字典的key是speaker2id

        # Load metadata of training data.
        metadata_path = Path(data_dir) / "metadata.json"
        metadata = json.load(open(metadata_path))["speakers"]

        # get the total number of speaker
        self.speaker_num = len(metadata.keys())
        self.data = []
        for speaker in metadata.keys():
            for utterance in metadata[speaker]:
                self.data.append([utterance["feature_path"], self.speaker2id
                                  [speaker]])

    def __len__(self):
        return len(self.data)

    def __getitem__(self, index):
        feature_path, speaker = self.data[index]
        # Load preprocessed mel-spectrogram.
        mel = torch.load(os.path.join(self.data_dir, feature_path))

        # segment mel-spectrogram into "segment_len" frames.
        if len(mel) >self.segment_len:
            start = random.randint(0, len(mel)-self.segment_len) #随便选取一个开始截取的位置 这个位置往后的长度要大于segment_len
            mel = torch.FloatTensor(mel[start:start+self.segment_len])
        else:
            mel = torch.FloatTensor(mel)

        # Turn the speaker id into long for computing loss later.
        speaker = torch.FloatTensor([speaker]).long

        return mel, speaker

    def get_speaker_number(self):
        return self.speaker_num

最后的数据中，发言人一共600个，语音一共69438个

4.DataLoader

DataLoader的完整参数列表

class torch.utils.data.DataLoader(
dataset,
batch_size=1,
shuffle=False,
sampler=None,
batch_sampler=None,
num_workers=0,
collate_fn=<function default_collate>,
pin_memory=False,
drop_last=False,
timeout=0,
worker_init_fn=None)

其中有一个参数collate_fn，可能比较陌生
collate_fn作用：
在最后一步堆叠的时候可能会出现问题: 如果一条数据中所含有的每个数据元的长度不同, 那么将无法进行堆叠. 如: multi-hot类型的数据, 序列数据。在使用这些数据时, 通常需要先进行长度上的补齐, 再进行堆叠. 以现在的流程, 是没有办法加入该操作的。此外, 某些优化方法是要对一个batch的数据进行操作。

collate_fn函数就是手动将抽取出的样本堆叠起来的函数。

所以我们需要自己定义collate_fn函数来统一特征大小

def collate_batch(batch):
    # Process features within a batch
    mel, speaker = zip(*batch)
    # Because we train the model batch by batch, we need to pad the features in the same batch to make their lengths the same
    mel = pad_sequence(mel, batch_first=True, padding_value=-20) # pad long 10^(-20) ehich is small value.
    # mel: (batch size, length, 40)
    return mel, torch.FloatTensor(speaker).long()

然后定义自己的dataloader，并在DataLoader中划分训练集和验证集

def get_dataloader(data_dir, batch_size, n_workers):
    dataset = myDataset(data_dir)
    speaker_num = dataset.get_speaker_number()
    # split
    trainlen = int(0.9 * len(dataset))
    lengths = [trainlen, len(dataset) - trainlen]
    trainset, validset = random_split(dataset, lengths)
    train_loader = DataLoader(
        trainset,
        batch_size=batch_size,
        shuffle=True,
        num_workers=n_workers,
        drop_last=True,
        pin_memory=True,
        collate_fn=collate_batch,
    )
    valid_loader = DataLoader(
        validset,
        batch_size=batch_size,
        num_workers=n_workers,
        drop_last=True,
        pin_memory=True,
        collate_fn=collate_batch,
    )
    return train_loader, valid_loader

5.Define Model

分类器是由transformerEncoder和全连接层构成的，输入是mels（shape为[batch size, length, 40]），输出是out（shape为[batch size, length, d_model]）

# -*- coding = utf-8 -*-
# @Time : 2023/4/7 14:44
# @Author : 头发没了还会再长
# @File : mdoel.py
# @Software : PyCharm
import torch
import torch.nn as nn
import torch.nn.functional as F
# 分类器 使用transformer
class Classifier(nn.Module):
    def __init__(self, d_model=80, n_spks=600, dropout=0.1):
        super().__init__()
        # project the dimession of features from that of input into d_model
        self.prenet = nn.Linear(40, d_model)
        self.encoder_layer = nn.TransformerEncoderLayer(
            d_model=d_model, dim_feadforeard=256, nhead=2
        )
        self.pred_layer = nn.Sequential(
            nn.Linear(d_model, d_model),
            nn.ReLU(),
            nn.Linear(d_model, n_spks),
        )

    def forward(self, mels):
        '''
        :param mels: (batch size, length, 40)
        :return: (batch size, length, d_model)
        '''
        out = self.prenet(mels)
        out = out.permute(1, 0, 2)
        out = self.encoder_layer(out)
        out = out.transpose(0, 1)
        stats = out.mean(dim=1)
        out = self.pred_layer(stats)
        return out

6.Train and Valid

6.1 learning rate schedule

The warmup schedule

• Set learning rate to 0 in the beginning
• The learning rate increases linearly from 0 to initial learning rate during warmup period

import math

import torch
from torch.optim import Optimizer
from torch.optim.lr_scheduler import LambdaLR


def get_cosine_schedule_with_warmup(
  optimizer: Optimizer,
  num_warmup_steps: int,
  num_training_steps: int,
  num_cycles: float = 0.5,
  last_epoch: int = -1,
):
  """
  Create a schedule with a learning rate that decreases following the values of the cosine function between the
  initial lr set in the optimizer to 0, after a warmup period during which it increases linearly between 0 and the
  initial lr set in the optimizer.

  Args:
    optimizer (:class:`~torch.optim.Optimizer`):
      The optimizer for which to schedule the learning rate.
    num_warmup_steps (:obj:`int`):
      The number of steps for the warmup phase.
    num_training_steps (:obj:`int`):
      The total number of training steps.
    num_cycles (:obj:`float`, `optional`, defaults to 0.5):
      The number of waves in the cosine schedule (the defaults is to just decrease from the max value to 0
      following a half-cosine).
    last_epoch (:obj:`int`, `optional`, defaults to -1):
      The index of the last epoch when resuming training.

  Return:
    :obj:`torch.optim.lr_scheduler.LambdaLR` with the appropriate schedule.
  """

  def lr_lambda(current_step):
    # Warmup
    if current_step < num_warmup_steps:
      return float(current_step) / float(max(1, num_warmup_steps))
    # decadence
    progress = float(current_step - num_warmup_steps) / float(
      max(1, num_training_steps - num_warmup_steps)
    )
    return max(
      0.0, 0.5 * (1.0 + math.cos(math.pi * float(num_cycles) * 2.0 * progress))
    )

  return LambdaLR(optimizer, lr_lambda, last_epoch)

6.2定义model_fn

输入一组batch，输出损失和准确率

def model_fn(batch, model, criterion, device):
    mels, labels = batch
    mels = mels.to(device)
    labels = labels.to(device)

    outs = model(mels)
    loss = criterion(outs, labels)

    preds = outs.argmax(1)
    accuracy = torch.mean((preds == labels).float())
    return loss, accuracy

6.3 validation

def valid(dataloader, model, criterion, device): 
  """Validate on validation set."""

  model.eval()
  running_loss = 0.0
  running_accuracy = 0.0
  pbar = tqdm(total=len(dataloader.dataset), ncols=0, desc="Valid", unit=" uttr")

  for i, batch in enumerate(dataloader):
    with torch.no_grad():
      loss, accuracy = model_fn(batch, model, criterion, device)
      running_loss += loss.item()
      running_accuracy += accuracy.item()

    pbar.update(dataloader.batch_size)
    pbar.set_postfix(
      loss=f"{running_loss / (i+1):.2f}",
      accuracy=f"{running_accuracy / (i+1):.2f}",
    )

  pbar.close()
  model.train()

  return running_accuracy / len(dataloader)

6.4 有了前期的准备，就可以开始训练模型了

• 首先，需要准备训练数据，即加载dataloader，定义损失函数，优化器等
• 然后开始循环，在训练集上计算梯度反向传播，在验证集上计算准确率
• 存储最优的模型

def parse_args():
  """arguments"""
  config = {
    "data_dir": "./Dataset",
    "save_path": "model.ckpt",
    "batch_size": 32,
    "n_workers": 8,
    "valid_steps": 2000,
    "warmup_steps": 1000,
    "save_steps": 10000,
    "total_steps": 70000,
  }

  return config


def main(
  data_dir,
  save_path,
  batch_size,
  n_workers,
  valid_steps,
  warmup_steps,
  total_steps,
  save_steps,
):
  """Main function."""
  device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
  print(f"[Info]: Use {device} now!")

  train_loader, valid_loader, speaker_num = get_dataloader(data_dir, batch_size, n_workers)
  train_iterator = iter(train_loader)
  print(f"[Info]: Finish loading data!",flush = True)

  model = Classifier(n_spks=speaker_num).to(device)
  criterion = nn.CrossEntropyLoss()
  optimizer = AdamW(model.parameters(), lr=1e-3)
  scheduler = get_cosine_schedule_with_warmup(optimizer, warmup_steps, total_steps)
  print(f"[Info]: Finish creating model!",flush = True)

  best_accuracy = -1.0
  best_state_dict = None

  pbar = tqdm(total=valid_steps, ncols=0, desc="Train", unit=" step")

  for step in range(total_steps):
    # Get data
    try:
      batch = next(train_iterator)
    except StopIteration:
      train_iterator = iter(train_loader)
      batch = next(train_iterator)

    loss, accuracy = model_fn(batch, model, criterion, device)
    batch_loss = loss.item()
    batch_accuracy = accuracy.item()

    # Updata model
    loss.backward()
    optimizer.step()
    scheduler.step()
    optimizer.zero_grad()
    
    # Log
    pbar.update()
    pbar.set_postfix(
      loss=f"{batch_loss:.2f}",
      accuracy=f"{batch_accuracy:.2f}",
      step=step + 1,
    )

    # Do validation
    if (step + 1) % valid_steps == 0:
      pbar.close()

      valid_accuracy = valid(valid_loader, model, criterion, device)

      # keep the best model
      if valid_accuracy > best_accuracy:
        best_accuracy = valid_accuracy
        best_state_dict = model.state_dict()

      pbar = tqdm(total=valid_steps, ncols=0, desc="Train", unit=" step")

    # Save the best model so far.
    if (step + 1) % save_steps == 0 and best_state_dict is not None:
      torch.save(best_state_dict, save_path)
      pbar.write(f"Step {step + 1}, best model saved. (accuracy={best_accuracy:.4f})")

  pbar.close()


if __name__ == "__main__":
  main(**parse_args())

这是训练过程中的截图：

7.Test

训练完模型，我们会得到最好的参数，并且保存在model.ckpt文件中了，接下来，只需要定义训练的函数，使用保存的最优模型开始训练即可。
测试和训练基本相同，需要准备dataset，dataloader

7.1 Inference dataset

class InferenceDataset(Dataset):
    def __init__(self, data_dir):
        testdata_path = Path(data_dir)
        metadata = json.load(testdata_path.open())
        self.data_dir = data_dir
        self.data = metadata["utterances"]

    def __len__(self):
        return len(self.data)

    def __getitem__(self, index):
        utterance = self.data[index]
        feature_path = utterance["feature_path"]
        mel = torch.load(os.path.join(self.data_dir, feature_path))

        return feature_path, mel

7.2可以开始预测啦

def inference_collate_batch(batch):
    feature_paths, mels = zip(*batch)
    return feature_paths, torch.stack(mels)

def parse_args():
  """arguments"""
  config = {
    "data_dir": "dataset",
    "model_path": "./model.ckpt",
    "output_path": "./output.csv",
  }

  return config

def main(data_dir,model_path,output_path,):
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    print(f"[Info]: Use {device} now!")

    mapping_path = Path(data_dir) / "mapping.json"
    mapping = json.load(mapping_path.open())

    dataset = InferenceDataset(data_dir)
    dataloader = DataLoader(
        dataset,
        batch_size=1,
        shuffle=False,
        drop_last=False,
        num_workers=0,
        collate_fn=inference_collate_batch,
    )
    print(f"[Info]: Finish loading data!", flush=True)

    speaker_num = len(mapping["id2speaker"])
    model = Classifier(n_spks=speaker_num).to(device)
    model.load_state_dict(torch.load(model_path))
    model.eval()
    print(f"[Info]: Finish creating model!", flush=True)
    results = [["Id", "Category"]]
    for feat_paths, mels in tqdm(dataloader):
        with torch.no_grad():
            mels = mels.to(device)
            outs = model(mels)
            preds = outs.argmax(1).cpu().numpy()
            for feat_path, pred in zip(feat_paths, preds):
                results.append([feat_path, mapping["id2speaker"][str(pred)]])

    with open(output_path, 'w', newline='') as csvfile:
        writer = csv.writer(csvfile)
        writer.writerows(results)