From 487600d0445dbd99e0bcb4e372943fea7b59ae12 Mon Sep 17 00:00:00 2001
From: Hung Dinh Xuan <hungdinhxdev@gmail.com>
Date: Sat, 21 Dec 2024 15:59:39 +0900
Subject: [PATCH] add configs for train private data

---
 .../normal_largecorpus_multiview_cnsl.yaml    |   1 +
 ...rmertcm_multiview_large_corpus_conf-2.yaml |  48 ++
 notebooks/analyze.ipynb                       | 700 +++++++++---------
 notebooks/eval.ipynb                          | 130 +++-
 notebooks/prepare_dataset.ipynb               | 185 ++++-
 src/data/components/augwrapper.py             |  20 +-
 src/data/normal_multiview_datamodule.py       |  19 +-
 7 files changed, 690 insertions(+), 413 deletions(-)
 create mode 100644 configs/experiment/xlsr_conformertcm_multiview_large_corpus_conf-2.yaml

diff --git a/configs/data/normal_largecorpus_multiview_cnsl.yaml b/configs/data/normal_largecorpus_multiview_cnsl.yaml
index 8f36302..621234c 100644
--- a/configs/data/normal_largecorpus_multiview_cnsl.yaml
+++ b/configs/data/normal_largecorpus_multiview_cnsl.yaml
@@ -29,6 +29,7 @@ args:
     augmentation_methods:
       ["RawBoost12", "pitch_1", "volume_10", "speed_01", "none"] # "none" is the original data
     wav_samp_rate: 16000
+    trim_length: 64000
     online_aug: true
     aug_dir: ${oc.env:LARGE_CORPUS_FOR_CNSL}/aug
     noise_path: ${oc.env:NOISE_PATH}
diff --git a/configs/experiment/xlsr_conformertcm_multiview_large_corpus_conf-2.yaml b/configs/experiment/xlsr_conformertcm_multiview_large_corpus_conf-2.yaml
new file mode 100644
index 0000000..68da92a
--- /dev/null
+++ b/configs/experiment/xlsr_conformertcm_multiview_large_corpus_conf-2.yaml
@@ -0,0 +1,48 @@
+# @package _global_
+
+# to execute this experiment run:
+# python train.py experiment=example
+
+defaults:
+  - override /data: normal_largecorpus_multiview_cnsl
+  - override /model: xlsr_conformertcm_baseline_multiview
+  - override /callbacks: default_loss
+  - override /trainer: default
+
+# all parameters below will be merged with parameters from default configurations set above
+# this allows you to overwrite only specified parameters
+
+tags: ["normal_largecorpus_multiview_cnsl", "xlsr_conformertcm_baseline_multiview"]
+
+seed: 1234
+
+trainer:
+  max_epochs: 100
+  gradient_clip_val: 0.0
+  accelerator: cuda
+
+model:
+  optimizer:
+    lr: 0.000001
+    weight_decay: 0.0001
+  net: null
+  scheduler: null
+  compile: true
+  cross_entropy_weight: [0.7, 0.3] # Balanced weights for cross entropy loss
+
+data:
+  batch_size: 14
+  num_workers: 8
+  pin_memory: true
+  args:
+    augmentation_methods: ["RawBoost12", 'RawBoost12', 'RawBoostFull', "pitch_1", "volume_10", "speed_01", "none"] # "none" is the original data
+    padding_type: repeat
+    random_start: True
+
+
+logger:
+  wandb:
+    tags: ${tags}
+    group: "normal_largecorpus_multiview_cnsl"
+  aim:
+    experiment: "normal_largecorpus_multiview_cnsl"
diff --git a/notebooks/analyze.ipynb b/notebooks/analyze.ipynb
index add77d0..8795665 100644
--- a/notebooks/analyze.ipynb
+++ b/notebooks/analyze.ipynb
@@ -1,353 +1,353 @@
 {
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Processing audio files:   0%|          | 0/31779 [00:00<?, ?it/s]/tmp/ipykernel_3393222/3850030267.py:28: FutureWarning: get_duration() keyword argument 'filename' has been renamed to 'path' in version 0.10.0.\n",
-      "\tThis alias will be removed in version 1.0.\n",
-      "  duration = librosa.get_duration(filename=audio_path)\n",
-      "Processing audio files: 100%|██████████| 31779/31779 [02:46<00:00, 190.63it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Results written to in_the_wild_durations.csv\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
+   "cells": [
+      {
+         "cell_type": "code",
+         "execution_count": 1,
+         "metadata": {},
+         "outputs": [
+            {
+               "name": "stderr",
+               "output_type": "stream",
+               "text": [
+                  "Processing audio files:   0%|          | 0/31779 [00:00<?, ?it/s]/tmp/ipykernel_3393222/3850030267.py:28: FutureWarning: get_duration() keyword argument 'filename' has been renamed to 'path' in version 0.10.0.\n",
+                  "\tThis alias will be removed in version 1.0.\n",
+                  "  duration = librosa.get_duration(filename=audio_path)\n",
+                  "Processing audio files: 100%|██████████| 31779/31779 [02:46<00:00, 190.63it/s]"
+               ]
+            },
+            {
+               "name": "stdout",
+               "output_type": "stream",
+               "text": [
+                  "Results written to in_the_wild_durations.csv\n"
+               ]
+            },
+            {
+               "name": "stderr",
+               "output_type": "stream",
+               "text": [
+                  "\n"
+               ]
+            }
+         ],
+         "source": [
+            "import os\n",
+            "import librosa\n",
+            "import csv\n",
+            "from tqdm import tqdm\n",
+            "\n",
+            "def get_audio_durations(audio_folder, output_csv):\n",
+            "    \"\"\"\n",
+            "    Calculate duration of each audio file in the folder and write results to CSV\n",
+            "    \n",
+            "    Parameters:\n",
+            "        audio_folder (str): Path to folder containing audio files\n",
+            "        output_csv (str): Path to output CSV file\n",
+            "    \"\"\"\n",
+            "    \n",
+            "    # Get list of audio files\n",
+            "    audio_files = [f for f in os.listdir(audio_folder) if f.endswith(('.wav', '.mp3', '.flac'))]\n",
+            "    \n",
+            "    # Open CSV file to write results\n",
+            "    with open(output_csv, 'w', newline='') as csvfile:\n",
+            "        writer = csv.writer(csvfile)\n",
+            "        writer.writerow(['filename', 'duration'])  # Write header\n",
+            "        \n",
+            "        # Process each audio file with progress bar\n",
+            "        for audio_file in tqdm(audio_files, desc=\"Processing audio files\"):\n",
+            "            try:\n",
+            "                # Load audio file and get duration\n",
+            "                audio_path = os.path.join(audio_folder, audio_file)\n",
+            "                duration = librosa.get_duration(filename=audio_path)\n",
+            "                \n",
+            "                # Write result to CSV\n",
+            "                writer.writerow([audio_file, f\"{duration:.2f}\"])\n",
+            "                \n",
+            "            except Exception as e:\n",
+            "                print(f\"Error processing {audio_file}: {str(e)}\")\n",
+            "\n",
+            "if __name__ == \"__main__\":\n",
+            "    # Example usage\n",
+            "    audio_folder = \"/data/hungdx/Lightning-hydra/data/in_the_wild\"\n",
+            "    output_csv = \"in_the_wild_durations.csv\"\n",
+            "    \n",
+            "    get_audio_durations(audio_folder, output_csv)\n",
+            "    print(f\"Results written to {output_csv}\")"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 3,
+         "metadata": {},
+         "outputs": [
+            {
+               "name": "stdout",
+               "output_type": "stream",
+               "text": [
+                  "Average duration:  4.287989552849366\n"
+               ]
+            }
+         ],
+         "source": [
+            "import pandas as pd\n",
+            "\n",
+            "df = pd.read_csv('in_the_wild_durations.csv')\n",
+            "\n",
+            "print(\"Average duration: \", df['duration'].mean())\n"
+         ]
+      },
+      {
+         "cell_type": "markdown",
+         "metadata": {},
+         "source": [
+            "# Check NaN input"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 5,
+         "metadata": {},
+         "outputs": [],
+         "source": [
+            "import sys\n",
+            "\n",
+            "sys.path.append('/data/hungdx/Lightning-hydra/src')\n",
+            "\n",
+            "from data.normal_multiview_datamodule import Dataset_for_dev\n",
+            "\n",
+            "\n",
+            "def genList(protocol_path, is_train=False, is_eval=False, is_dev=False):\n",
+            "    \"\"\"\n",
+            "            This function generates the list of files and their corresponding labels\n",
+            "            Specifically for the standard CNSL dataset\n",
+            "        \"\"\"\n",
+            "     # bonafide: 1, spoof: 0\n",
+            "    d_meta = {}\n",
+            "    file_list = []\n",
+            "\n",
+            "    if (is_train):\n",
+            "        with open(protocol_path, 'r') as f:\n",
+            "            l_meta = f.readlines()\n",
+            "        for line in l_meta:\n",
+            "            utt, subset, label = line.strip().split()\n",
+            "            if subset == 'train':\n",
+            "                file_list.append(utt)\n",
+            "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
+            "\n",
+            "        return d_meta, file_list\n",
+            "    if (is_dev):\n",
+            "        with open(protocol_path, 'r') as f:\n",
+            "            l_meta = f.readlines()\n",
+            "        for line in l_meta:\n",
+            "            utt, subset, label = line.strip().split()\n",
+            "            if subset == 'dev':\n",
+            "                file_list.append(utt)\n",
+            "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
+            "        return d_meta, file_list\n",
+            "\n",
+            "    if (is_eval):\n",
+            "        # no eval protocol_path yet\n",
+            "        with open(protocol_path, 'r') as f:\n",
+            "            l_meta = f.readlines()\n",
+            "        for line in l_meta:\n",
+            "            utt, subset, label = line.strip().split()\n",
+            "            if subset == 'eval':\n",
+            "                file_list.append(utt)\n",
+            "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
+            "        # return d_meta, file_list\n",
+            "        return d_meta, file_list"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 21,
+         "metadata": {},
+         "outputs": [],
+         "source": [
+            "from dataclasses import dataclass\n",
+            "from dataclasses import dataclass, asdict\n",
+            "from typing import List, Optional\n",
+            "\n",
+            "\n",
+            "@dataclass\n",
+            "class DataConfig:\n",
+            "    augmentation_methods: List[str]\n",
+            "    wav_samp_rate: int\n",
+            "    online_aug: bool\n",
+            "    aug_dir: str\n",
+            "    noise_path: str\n",
+            "    rir_path: str\n",
+            "    repeat_pad: bool\n",
+            "    random_start: bool\n",
+            "\n",
+            "\n",
+            "@dataclass\n",
+            "class Args:\n",
+            "    views: List[int]\n",
+            "    protocol_path: str\n",
+            "    nBands: int\n",
+            "    minF: int\n",
+            "    maxF: int\n",
+            "    minBW: int\n",
+            "    maxBW: int\n",
+            "    minCoeff: int\n",
+            "    maxCoeff: int\n",
+            "    minG: int\n",
+            "    maxG: int\n",
+            "    minBiasLinNonLin: int\n",
+            "    maxBiasLinNonLin: int\n",
+            "    N_f: int\n",
+            "    P: int\n",
+            "    g_sd: int\n",
+            "    SNRmin: int\n",
+            "    SNRmax: int\n",
+            "    data: DataConfig\n",
+            "\n",
+            "    def __getitem__(self, key):\n",
+            "        if key == 'data':\n",
+            "            return asdict(self.data)\n",
+            "        return asdict(self)[key]\n",
+            "\n",
+            "    @classmethod\n",
+            "    def from_dict(cls, config_dict: dict):\n",
+            "        data_config = DataConfig(**config_dict['data'])\n",
+            "        config_dict['data'] = data_config\n",
+            "        return cls(**config_dict)\n",
+            "\n",
+            "\n",
+            "# Example usage:\n",
+            "config_dict = {\n",
+            "    'views': [1, 2, 3, 4],\n",
+            "    'protocol_path': '${oc.env:LARGE_CORPUS_FOR_CNSL_PROTOCOLS}',\n",
+            "    'nBands': 5,\n",
+            "    'minF': 20,\n",
+            "    'maxF': 8000,\n",
+            "    'minBW': 100,\n",
+            "    'maxBW': 1000,\n",
+            "    'minCoeff': 10,\n",
+            "    'maxCoeff': 100,\n",
+            "    'minG': 0,\n",
+            "    'maxG': 0,\n",
+            "    'minBiasLinNonLin': 5,\n",
+            "    'maxBiasLinNonLin': 20,\n",
+            "    'N_f': 5,\n",
+            "    'P': 10,\n",
+            "    'g_sd': 2,\n",
+            "    'SNRmin': 10,\n",
+            "    'SNRmax': 40,\n",
+            "    'data': {\n",
+            "        'augmentation_methods': [\"RawBoost12\", \"pitch_1\", \"volume_10\", \"speed_01\", \"none\"],\n",
+            "        'wav_samp_rate': 16000,\n",
+            "        'online_aug': True,\n",
+            "        'aug_dir': '${oc.env:LARGE_CORPUS_FOR_CNSL}/aug',\n",
+            "        'noise_path': '${oc.env:NOISE_PATH}',\n",
+            "        'rir_path': '${oc.env:RIR_PATH}',\n",
+            "        'repeat_pad': True,\n",
+            "        'random_start': True\n",
+            "    }\n",
+            "}\n",
+            "# Create Args object\n",
+            "args = Args.from_dict(config_dict)"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 23,
+         "metadata": {},
+         "outputs": [
+            {
+               "name": "stdout",
+               "output_type": "stream",
+               "text": [
+                  "vocoders: []\n"
+               ]
+            }
+         ],
+         "source": [
+            "protocol_path = \"/data/hungdx/Lightning-hydra/notebooks/new_protocol_trim_vocoded_cleaned.txt\"\n",
+            "data_dir = \"/data/hungdx/Lightning-hydra/data/0_large-corpus\"\n",
+            "\n",
+            "d_label_dev, file_dev = genList(protocol_path,\n",
+            "    is_train=False, is_eval=False, is_dev=True)\n",
+            "\n",
+            "data_val = Dataset_for_dev(args, list_IDs=file_dev, labels=d_label_dev,\n",
+            "                       base_dir=data_dir+'/',  is_train=False, **args['data'])"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 27,
+         "metadata": {},
+         "outputs": [],
+         "source": [
+            "from torch.utils.data import DataLoader, Dataset\n",
+            "from data.components.collate_fn import multi_view_collate_fn, variable_multi_view_collate_fn\n",
+            "\n",
+            "\n",
+            "def collate_fn(x): return multi_view_collate_fn(\n",
+            "    x,\n",
+            "    args.views,\n",
+            "    args.data.wav_samp_rate,\n",
+            "    \"repeat\",\n",
+            "    True\n",
+            ")\n",
+            "\n",
+            "dev_dataloader = DataLoader(\n",
+            "    dataset=data_val,\n",
+            "    batch_size=1,\n",
+            "    num_workers=1,\n",
+            "    pin_memory=True,\n",
+            "    shuffle=False,\n",
+            "    collate_fn=collate_fn,\n",
+            ")"
+         ]
+      },
+      {
+         "cell_type": "code",
+         "execution_count": 30,
+         "metadata": {},
+         "outputs": [
+            {
+               "name": "stderr",
+               "output_type": "stream",
+               "text": [
+                  "Validating: 100%|██████████| 66725/66725 [08:32<00:00, 130.08it/s]\n"
+               ]
+            }
+         ],
+         "source": [
+            "import torch\n",
+            "from tqdm import tqdm\n",
+            "\n",
+            "# Iterate through the DataLoader with a progress bar\n",
+            "for batch_idx, batch in enumerate(tqdm(dev_dataloader, desc=\"Validating\")):\n",
+            "    for view, (x, y) in batch.items():\n",
+            "        if torch.isnan(x).any() or torch.isinf(x).any():\n",
+            "            print(\"Found NaN or Inf in validation inputs!\")\n",
+            "        if torch.isnan(y).any() or torch.isinf(y).any():\n",
+            "            print(\"Found NaN or Inf in validation labels!\")"
+         ]
+      }
    ],
-   "source": [
-    "import os\n",
-    "import librosa\n",
-    "import csv\n",
-    "from tqdm import tqdm\n",
-    "\n",
-    "def get_audio_durations(audio_folder, output_csv):\n",
-    "    \"\"\"\n",
-    "    Calculate duration of each audio file in the folder and write results to CSV\n",
-    "    \n",
-    "    Parameters:\n",
-    "        audio_folder (str): Path to folder containing audio files\n",
-    "        output_csv (str): Path to output CSV file\n",
-    "    \"\"\"\n",
-    "    \n",
-    "    # Get list of audio files\n",
-    "    audio_files = [f for f in os.listdir(audio_folder) if f.endswith(('.wav', '.mp3', '.flac'))]\n",
-    "    \n",
-    "    # Open CSV file to write results\n",
-    "    with open(output_csv, 'w', newline='') as csvfile:\n",
-    "        writer = csv.writer(csvfile)\n",
-    "        writer.writerow(['filename', 'duration'])  # Write header\n",
-    "        \n",
-    "        # Process each audio file with progress bar\n",
-    "        for audio_file in tqdm(audio_files, desc=\"Processing audio files\"):\n",
-    "            try:\n",
-    "                # Load audio file and get duration\n",
-    "                audio_path = os.path.join(audio_folder, audio_file)\n",
-    "                duration = librosa.get_duration(filename=audio_path)\n",
-    "                \n",
-    "                # Write result to CSV\n",
-    "                writer.writerow([audio_file, f\"{duration:.2f}\"])\n",
-    "                \n",
-    "            except Exception as e:\n",
-    "                print(f\"Error processing {audio_file}: {str(e)}\")\n",
-    "\n",
-    "if __name__ == \"__main__\":\n",
-    "    # Example usage\n",
-    "    audio_folder = \"/data/hungdx/Lightning-hydra/data/in_the_wild\"\n",
-    "    output_csv = \"in_the_wild_durations.csv\"\n",
-    "    \n",
-    "    get_audio_durations(audio_folder, output_csv)\n",
-    "    print(f\"Results written to {output_csv}\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Average duration:  4.287989552849366\n"
-     ]
-    }
-   ],
-   "source": [
-    "import pandas as pd\n",
-    "\n",
-    "df = pd.read_csv('in_the_wild_durations.csv')\n",
-    "\n",
-    "print(\"Average duration: \", df['duration'].mean())\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Check NaN input"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import sys\n",
-    "\n",
-    "sys.path.append('/data/hungdx/Lightning-hydra/src')\n",
-    "\n",
-    "from data.normal_multiview_datamodule import Dataset_for_dev\n",
-    "\n",
-    "\n",
-    "def genList(protocol_path, is_train=False, is_eval=False, is_dev=False):\n",
-    "    \"\"\"\n",
-    "            This function generates the list of files and their corresponding labels\n",
-    "            Specifically for the standard CNSL dataset\n",
-    "        \"\"\"\n",
-    "     # bonafide: 1, spoof: 0\n",
-    "    d_meta = {}\n",
-    "    file_list = []\n",
-    "\n",
-    "    if (is_train):\n",
-    "        with open(protocol_path, 'r') as f:\n",
-    "            l_meta = f.readlines()\n",
-    "        for line in l_meta:\n",
-    "            utt, subset, label = line.strip().split()\n",
-    "            if subset == 'train':\n",
-    "                file_list.append(utt)\n",
-    "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
-    "\n",
-    "        return d_meta, file_list\n",
-    "    if (is_dev):\n",
-    "        with open(protocol_path, 'r') as f:\n",
-    "            l_meta = f.readlines()\n",
-    "        for line in l_meta:\n",
-    "            utt, subset, label = line.strip().split()\n",
-    "            if subset == 'dev':\n",
-    "                file_list.append(utt)\n",
-    "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
-    "        return d_meta, file_list\n",
-    "\n",
-    "    if (is_eval):\n",
-    "        # no eval protocol_path yet\n",
-    "        with open(protocol_path, 'r') as f:\n",
-    "            l_meta = f.readlines()\n",
-    "        for line in l_meta:\n",
-    "            utt, subset, label = line.strip().split()\n",
-    "            if subset == 'eval':\n",
-    "                file_list.append(utt)\n",
-    "                d_meta[utt] = 1 if label == 'bonafide' else 0\n",
-    "        # return d_meta, file_list\n",
-    "        return d_meta, file_list"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from dataclasses import dataclass\n",
-    "from dataclasses import dataclass, asdict\n",
-    "from typing import List, Optional\n",
-    "\n",
-    "\n",
-    "@dataclass\n",
-    "class DataConfig:\n",
-    "    augmentation_methods: List[str]\n",
-    "    wav_samp_rate: int\n",
-    "    online_aug: bool\n",
-    "    aug_dir: str\n",
-    "    noise_path: str\n",
-    "    rir_path: str\n",
-    "    repeat_pad: bool\n",
-    "    random_start: bool\n",
-    "\n",
-    "\n",
-    "@dataclass\n",
-    "class Args:\n",
-    "    views: List[int]\n",
-    "    protocol_path: str\n",
-    "    nBands: int\n",
-    "    minF: int\n",
-    "    maxF: int\n",
-    "    minBW: int\n",
-    "    maxBW: int\n",
-    "    minCoeff: int\n",
-    "    maxCoeff: int\n",
-    "    minG: int\n",
-    "    maxG: int\n",
-    "    minBiasLinNonLin: int\n",
-    "    maxBiasLinNonLin: int\n",
-    "    N_f: int\n",
-    "    P: int\n",
-    "    g_sd: int\n",
-    "    SNRmin: int\n",
-    "    SNRmax: int\n",
-    "    data: DataConfig\n",
-    "\n",
-    "    def __getitem__(self, key):\n",
-    "        if key == 'data':\n",
-    "            return asdict(self.data)\n",
-    "        return asdict(self)[key]\n",
-    "\n",
-    "    @classmethod\n",
-    "    def from_dict(cls, config_dict: dict):\n",
-    "        data_config = DataConfig(**config_dict['data'])\n",
-    "        config_dict['data'] = data_config\n",
-    "        return cls(**config_dict)\n",
-    "\n",
-    "\n",
-    "# Example usage:\n",
-    "config_dict = {\n",
-    "    'views': [1, 2, 3, 4],\n",
-    "    'protocol_path': '${oc.env:LARGE_CORPUS_FOR_CNSL_PROTOCOLS}',\n",
-    "    'nBands': 5,\n",
-    "    'minF': 20,\n",
-    "    'maxF': 8000,\n",
-    "    'minBW': 100,\n",
-    "    'maxBW': 1000,\n",
-    "    'minCoeff': 10,\n",
-    "    'maxCoeff': 100,\n",
-    "    'minG': 0,\n",
-    "    'maxG': 0,\n",
-    "    'minBiasLinNonLin': 5,\n",
-    "    'maxBiasLinNonLin': 20,\n",
-    "    'N_f': 5,\n",
-    "    'P': 10,\n",
-    "    'g_sd': 2,\n",
-    "    'SNRmin': 10,\n",
-    "    'SNRmax': 40,\n",
-    "    'data': {\n",
-    "        'augmentation_methods': [\"RawBoost12\", \"pitch_1\", \"volume_10\", \"speed_01\", \"none\"],\n",
-    "        'wav_samp_rate': 16000,\n",
-    "        'online_aug': True,\n",
-    "        'aug_dir': '${oc.env:LARGE_CORPUS_FOR_CNSL}/aug',\n",
-    "        'noise_path': '${oc.env:NOISE_PATH}',\n",
-    "        'rir_path': '${oc.env:RIR_PATH}',\n",
-    "        'repeat_pad': True,\n",
-    "        'random_start': True\n",
-    "    }\n",
-    "}\n",
-    "# Create Args object\n",
-    "args = Args.from_dict(config_dict)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "vocoders: []\n"
-     ]
-    }
-   ],
-   "source": [
-    "protocol_path = \"/data/hungdx/Lightning-hydra/notebooks/new_protocol_trim_vocoded_cleaned.txt\"\n",
-    "data_dir = \"/data/hungdx/Lightning-hydra/data/0_large-corpus\"\n",
-    "\n",
-    "d_label_dev, file_dev = genList(protocol_path,\n",
-    "    is_train=False, is_eval=False, is_dev=True)\n",
-    "\n",
-    "data_val = Dataset_for_dev(args, list_IDs=file_dev, labels=d_label_dev,\n",
-    "                       base_dir=data_dir+'/',  is_train=False, **args['data'])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from torch.utils.data import DataLoader, Dataset\n",
-    "from data.components.collate_fn import multi_view_collate_fn, variable_multi_view_collate_fn\n",
-    "\n",
-    "\n",
-    "def collate_fn(x): return multi_view_collate_fn(\n",
-    "    x,\n",
-    "    args.views,\n",
-    "    args.data.wav_samp_rate,\n",
-    "    \"repeat\",\n",
-    "    True\n",
-    ")\n",
-    "\n",
-    "dev_dataloader = DataLoader(\n",
-    "    dataset=data_val,\n",
-    "    batch_size=1,\n",
-    "    num_workers=1,\n",
-    "    pin_memory=True,\n",
-    "    shuffle=False,\n",
-    "    collate_fn=collate_fn,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Validating: 100%|██████████| 66725/66725 [08:32<00:00, 130.08it/s]\n"
-     ]
-    }
-   ],
-   "source": [
-    "import torch\n",
-    "from tqdm import tqdm\n",
-    "\n",
-    "# Iterate through the DataLoader with a progress bar\n",
-    "for batch_idx, batch in enumerate(tqdm(dev_dataloader, desc=\"Validating\")):\n",
-    "    for view, (x, y) in batch.items():\n",
-    "        if torch.isnan(x).any() or torch.isinf(x).any():\n",
-    "            print(\"Found NaN or Inf in validation inputs!\")\n",
-    "        if torch.isnan(y).any() or torch.isinf(y).any():\n",
-    "            print(\"Found NaN or Inf in validation labels!\")"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "asvspoof5",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
+   "metadata": {
+      "kernelspec": {
+         "display_name": "asvspoof5",
+         "language": "python",
+         "name": "python3"
+      },
+      "language_info": {
+         "codemirror_mode": {
+            "name": "ipython",
+            "version": 3
+         },
+         "file_extension": ".py",
+         "mimetype": "text/x-python",
+         "name": "python",
+         "nbconvert_exporter": "python",
+         "pygments_lexer": "ipython3",
+         "version": "3.9.19"
+      }
    },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.9.19"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}
+   "nbformat": 4,
+   "nbformat_minor": 2
+}
\ No newline at end of file
diff --git a/notebooks/eval.ipynb b/notebooks/eval.ipynb
index 5b4328d..fad9685 100644
--- a/notebooks/eval.ipynb
+++ b/notebooks/eval.ipynb
@@ -103,31 +103,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "/data/hungdx/Lightning-hydra/logs/aasist_ssl_baseline_scores/inthewild/eval_CM_scores_file_SSL_DF_1s.txt\n"
-     ]
-    },
-    {
-     "ename": "IndexError",
-     "evalue": "index 0 is out of bounds for axis 0 with size 0",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[4], line 40\u001b[0m\n\u001b[1;32m     38\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m score_file \u001b[38;5;129;01min\u001b[39;00m list_dir:\n\u001b[1;32m     39\u001b[0m     \u001b[38;5;28mprint\u001b[39m(score_file)\n\u001b[0;32m---> 40\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\u001b[43meval_to_score_file\u001b[49m\u001b[43m(\u001b[49m\u001b[43mscore_file\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m/datad/Datasets/in_the_wild.txt\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m)\n\u001b[1;32m     41\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
-      "Cell \u001b[0;32mIn[4], line 25\u001b[0m, in \u001b[0;36meval_to_score_file\u001b[0;34m(score_file, cm_key_file)\u001b[0m\n\u001b[1;32m     22\u001b[0m bona_cm \u001b[38;5;241m=\u001b[39m cm_scores[cm_scores[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m1_y\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mbonafide\u001b[39m\u001b[38;5;124m'\u001b[39m][\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m1_x\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m.\u001b[39mvalues\n\u001b[1;32m     23\u001b[0m spoof_cm \u001b[38;5;241m=\u001b[39m cm_scores[cm_scores[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m1_y\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mspoof\u001b[39m\u001b[38;5;124m'\u001b[39m][\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m1_x\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m.\u001b[39mvalues\n\u001b[0;32m---> 25\u001b[0m eer_cm, th \u001b[38;5;241m=\u001b[39m \u001b[43mem\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcompute_eer\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbona_cm\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mspoof_cm\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     26\u001b[0m out_data \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124meer: \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;130;01m\\t\u001b[39;00m\u001b[38;5;124mthreshold: \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(\u001b[38;5;241m100\u001b[39m\u001b[38;5;241m*\u001b[39meer_cm, th)\n\u001b[1;32m     27\u001b[0m \u001b[38;5;28mprint\u001b[39m(out_data)\n",
-      "File \u001b[0;32m/data/hungdx/Lightning-hydra/notebooks/eval_metrics_DF.py:44\u001b[0m, in \u001b[0;36mcompute_eer\u001b[0;34m(target_scores, nontarget_scores)\u001b[0m\n\u001b[1;32m     42\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mcompute_eer\u001b[39m(target_scores, nontarget_scores):\n\u001b[1;32m     43\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124;03m\"\"\" Returns equal error rate (EER) and the corresponding threshold. \"\"\"\u001b[39;00m\n\u001b[0;32m---> 44\u001b[0m     frr, far, thresholds \u001b[38;5;241m=\u001b[39m \u001b[43mcompute_det_curve\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtarget_scores\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnontarget_scores\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     45\u001b[0m     abs_diffs \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mabs(frr \u001b[38;5;241m-\u001b[39m far)\n\u001b[1;32m     46\u001b[0m     min_index \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39margmin(abs_diffs)\n",
-      "File \u001b[0;32m/data/hungdx/Lightning-hydra/notebooks/eval_metrics_DF.py:37\u001b[0m, in \u001b[0;36mcompute_det_curve\u001b[0;34m(target_scores, nontarget_scores)\u001b[0m\n\u001b[1;32m     35\u001b[0m frr \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mconcatenate((np\u001b[38;5;241m.\u001b[39matleast_1d(\u001b[38;5;241m0\u001b[39m), tar_trial_sums \u001b[38;5;241m/\u001b[39m target_scores\u001b[38;5;241m.\u001b[39msize))  \u001b[38;5;66;03m# false rejection rates\u001b[39;00m\n\u001b[1;32m     36\u001b[0m far \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mconcatenate((np\u001b[38;5;241m.\u001b[39matleast_1d(\u001b[38;5;241m1\u001b[39m), nontarget_trial_sums \u001b[38;5;241m/\u001b[39m nontarget_scores\u001b[38;5;241m.\u001b[39msize))  \u001b[38;5;66;03m# false acceptance rates\u001b[39;00m\n\u001b[0;32m---> 37\u001b[0m thresholds \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mconcatenate((np\u001b[38;5;241m.\u001b[39matleast_1d(all_scores[\u001b[43mindices\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m]\u001b[49m] \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m0.001\u001b[39m), all_scores[indices]))  \u001b[38;5;66;03m# Thresholds are the sorted scores\u001b[39;00m\n\u001b[1;32m     39\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m frr, far, thresholds\n",
-      "\u001b[0;31mIndexError\u001b[0m: index 0 is out of bounds for axis 0 with size 0"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import sys\n",
     "import os.path\n",
@@ -174,16 +152,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "eer: 8.234916170745263\tthreshold: -3.04897689819336\n",
-      "\n",
-      "0.08234916170745263\n"
+     "ename": "FileNotFoundError",
+     "evalue": "[Errno 2] No such file or directory: '/dataa/Datasets/in_the_wild.txt'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[3], line 29\u001b[0m\n\u001b[1;32m     26\u001b[0m     \u001b[38;5;28mprint\u001b[39m(out_data)\n\u001b[1;32m     27\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m eer_cm\n\u001b[0;32m---> 29\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[43meval_to_score_file\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m/data/hungdx/tcm_add/Scores/avg_5_best_baseline_itw_var.txt\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m/dataa/Datasets/in_the_wild.txt\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m)\n",
+      "Cell \u001b[0;32mIn[3], line 13\u001b[0m, in \u001b[0;36meval_to_score_file\u001b[0;34m(score_file, cm_key_file)\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21meval_to_score_file\u001b[39m(score_file, cm_key_file):\n\u001b[1;32m      9\u001b[0m     \u001b[38;5;66;03m# CM key file is the metadata file that contains the ground truth labels for the eval set\u001b[39;00m\n\u001b[1;32m     10\u001b[0m     \u001b[38;5;66;03m# score file is the output of the system that contains the scores for the eval set\u001b[39;00m\n\u001b[1;32m     11\u001b[0m     \u001b[38;5;66;03m# phase is the phase of the eval set (dev or eval)\u001b[39;00m\n\u001b[0;32m---> 13\u001b[0m     cm_data \u001b[38;5;241m=\u001b[39m \u001b[43mpandas\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread_csv\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcm_key_file\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43msep\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43m \u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mheader\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[1;32m     14\u001b[0m     submission_scores \u001b[38;5;241m=\u001b[39m pandas\u001b[38;5;241m.\u001b[39mread_csv(\n\u001b[1;32m     15\u001b[0m         score_file, sep\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m \u001b[39m\u001b[38;5;124m'\u001b[39m, header\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, skipinitialspace\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m     16\u001b[0m     \u001b[38;5;66;03m# check here for progress vs eval set\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.11/site-packages/pandas/io/parsers/readers.py:948\u001b[0m, in \u001b[0;36mread_csv\u001b[0;34m(filepath_or_buffer, sep, delimiter, header, names, index_col, usecols, dtype, engine, converters, true_values, false_values, skipinitialspace, skiprows, skipfooter, nrows, na_values, keep_default_na, na_filter, verbose, skip_blank_lines, parse_dates, infer_datetime_format, keep_date_col, date_parser, date_format, dayfirst, cache_dates, iterator, chunksize, compression, thousands, decimal, lineterminator, quotechar, quoting, doublequote, escapechar, comment, encoding, encoding_errors, dialect, on_bad_lines, delim_whitespace, low_memory, memory_map, float_precision, storage_options, dtype_backend)\u001b[0m\n\u001b[1;32m    935\u001b[0m kwds_defaults \u001b[38;5;241m=\u001b[39m _refine_defaults_read(\n\u001b[1;32m    936\u001b[0m     dialect,\n\u001b[1;32m    937\u001b[0m     delimiter,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    944\u001b[0m     dtype_backend\u001b[38;5;241m=\u001b[39mdtype_backend,\n\u001b[1;32m    945\u001b[0m )\n\u001b[1;32m    946\u001b[0m kwds\u001b[38;5;241m.\u001b[39mupdate(kwds_defaults)\n\u001b[0;32m--> 948\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_read\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfilepath_or_buffer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.11/site-packages/pandas/io/parsers/readers.py:611\u001b[0m, in \u001b[0;36m_read\u001b[0;34m(filepath_or_buffer, kwds)\u001b[0m\n\u001b[1;32m    608\u001b[0m _validate_names(kwds\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnames\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m    610\u001b[0m \u001b[38;5;66;03m# Create the parser.\u001b[39;00m\n\u001b[0;32m--> 611\u001b[0m parser \u001b[38;5;241m=\u001b[39m \u001b[43mTextFileReader\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfilepath_or_buffer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    613\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m chunksize \u001b[38;5;129;01mor\u001b[39;00m iterator:\n\u001b[1;32m    614\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m parser\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.11/site-packages/pandas/io/parsers/readers.py:1448\u001b[0m, in \u001b[0;36mTextFileReader.__init__\u001b[0;34m(self, f, engine, **kwds)\u001b[0m\n\u001b[1;32m   1445\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39moptions[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhas_index_names\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m kwds[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhas_index_names\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[1;32m   1447\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhandles: IOHandles \u001b[38;5;241m|\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m-> 1448\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_engine \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_make_engine\u001b[49m\u001b[43m(\u001b[49m\u001b[43mf\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mengine\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.11/site-packages/pandas/io/parsers/readers.py:1705\u001b[0m, in \u001b[0;36mTextFileReader._make_engine\u001b[0;34m(self, f, engine)\u001b[0m\n\u001b[1;32m   1703\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m mode:\n\u001b[1;32m   1704\u001b[0m         mode \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m-> 1705\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhandles \u001b[38;5;241m=\u001b[39m \u001b[43mget_handle\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1706\u001b[0m \u001b[43m    \u001b[49m\u001b[43mf\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1707\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmode\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1708\u001b[0m \u001b[43m    \u001b[49m\u001b[43mencoding\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moptions\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mencoding\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1709\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcompression\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moptions\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mcompression\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1710\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmemory_map\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moptions\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mmemory_map\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1711\u001b[0m \u001b[43m    \u001b[49m\u001b[43mis_text\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mis_text\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1712\u001b[0m \u001b[43m    \u001b[49m\u001b[43merrors\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moptions\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mencoding_errors\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mstrict\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1713\u001b[0m \u001b[43m    \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43moptions\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mstorage_options\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1714\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1715\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhandles \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m   1716\u001b[0m f \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhandles\u001b[38;5;241m.\u001b[39mhandle\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.11/site-packages/pandas/io/common.py:863\u001b[0m, in \u001b[0;36mget_handle\u001b[0;34m(path_or_buf, mode, encoding, compression, memory_map, is_text, errors, storage_options)\u001b[0m\n\u001b[1;32m    858\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(handle, \u001b[38;5;28mstr\u001b[39m):\n\u001b[1;32m    859\u001b[0m     \u001b[38;5;66;03m# Check whether the filename is to be opened in binary mode.\u001b[39;00m\n\u001b[1;32m    860\u001b[0m     \u001b[38;5;66;03m# Binary mode does not support 'encoding' and 'newline'.\u001b[39;00m\n\u001b[1;32m    861\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m ioargs\u001b[38;5;241m.\u001b[39mencoding \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m ioargs\u001b[38;5;241m.\u001b[39mmode:\n\u001b[1;32m    862\u001b[0m         \u001b[38;5;66;03m# Encoding\u001b[39;00m\n\u001b[0;32m--> 863\u001b[0m         handle \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mopen\u001b[39m(\n\u001b[1;32m    864\u001b[0m             handle,\n\u001b[1;32m    865\u001b[0m             ioargs\u001b[38;5;241m.\u001b[39mmode,\n\u001b[1;32m    866\u001b[0m             encoding\u001b[38;5;241m=\u001b[39mioargs\u001b[38;5;241m.\u001b[39mencoding,\n\u001b[1;32m    867\u001b[0m             errors\u001b[38;5;241m=\u001b[39merrors,\n\u001b[1;32m    868\u001b[0m             newline\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m    869\u001b[0m         )\n\u001b[1;32m    870\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    871\u001b[0m         \u001b[38;5;66;03m# Binary mode\u001b[39;00m\n\u001b[1;32m    872\u001b[0m         handle \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mopen\u001b[39m(handle, ioargs\u001b[38;5;241m.\u001b[39mmode)\n",
+      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '/dataa/Datasets/in_the_wild.txt'"
      ]
     }
    ],
@@ -1960,15 +1946,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "/home/hungdx/mlaad_xlsr_conformertcm_train_large_corpus_multiview_conf-1.txt\n",
-      "Accuracy: 93.54\n",
+      "/home/hungdx/mlaad_xlsr_conformertcm_train_large_corpus_multiview_conf-1_lts_3s.txt\n",
+      "Accuracy: 93.48\n",
       "\n",
       "\n"
      ]
@@ -1990,7 +1976,7 @@
     "# list_dir = sorted(list_dir)\n",
     "\n",
     "#for score_file in list_dir:\n",
-    "score_file = \"/home/hungdx/mlaad_xlsr_conformertcm_train_large_corpus_multiview_conf-1.txt\"\n",
+    "score_file = \"/home/hungdx/mlaad_xlsr_conformertcm_train_large_corpus_multiview_conf-1_lts_3s.txt\"\n",
     "pred_df = pd.read_csv(score_file, sep=\" \", header=None)\n",
     "print(score_file)\n",
     "pred_df.columns = [\"utt\", \"spoof\", \"score\"]\n",
@@ -2011,16 +1997,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "eer: 0.22422491216022286\tthreshold: -0.7054389119148254\n",
+      "eer: 0.254385417374277\tthreshold: -0.7286882996559143\n",
       "\n",
-      "0.0022422491216022287\n",
+      "0.00254385417374277\n",
       "\n",
       "\n"
      ]
@@ -2062,10 +2048,76 @@
     "    return eer_cm\n",
     "\n",
     "\n",
-    "score_file = \"/home/hungdx/large_corpus_eval_xlsr_conformertcm_train_large_corpus_multiview_conf-1.txt\"\n",
+    "score_file = \"/home/hungdx/large_corpus_xlsr_conformertcm_train_large_corpus_multiview_conf-1_lts_2s.txt\"\n",
     "print(eval_to_score_file(score_file, \"/data/Datasets/0_large-corpus/protocol.txt\"))\n",
     "print(\"\\n\")"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# MLAAD_v5"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## config1 + Nov "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "/home/hungdx/mlaad_v5_xlsr_conformertcm_train_large_corpus_multiview_conf-1_lts_4s.txt\n",
+      "Accuracy: 95.18\n",
+      "\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, accuracy_score, f1_score, recall_score, precision_score, det_curve\n",
+    "full_df = pd.read_csv(\n",
+    "    \"/nvme1/hungdx/Lightning-hydra/data/mlaad_v5/mlaad_v5_protocol.txt\", sep=\" \", header=None)\n",
+    "\n",
+    "full_df.columns = [\"utt\", \"subset\", \"label\"]\n",
+    "\n",
+    "# BASE_DIR = \"/nvme2/hungdx/tcm_add/Scores/CNSL_Multiview/ConformerTCM_LargeCorpus_eval_MLAAD\"\n",
+    "# list_dir = os.listdir(\n",
+    "#     BASE_DIR)\n",
+    "\n",
+    "# list_dir = [os.path.join(BASE_DIR, x) for x in list_dir]\n",
+    "# list_dir = sorted(list_dir)\n",
+    "\n",
+    "# for score_file in list_dir:\n",
+    "score_file = \"/home/hungdx/mlaad_v5_xlsr_conformertcm_train_large_corpus_multiview_conf-1_lts_4s.txt\"\n",
+    "pred_df = pd.read_csv(score_file, sep=\" \", header=None)\n",
+    "print(score_file)\n",
+    "pred_df.columns = [\"utt\", \"spoof\", \"score\"]\n",
+    "pred_df = pred_df.drop_duplicates(subset=['utt'])\n",
+    "\n",
+    "# pred_df['utt'] = pred_df['utt'].apply(lambda x: x.split('/')[-1].split('.')[0])\n",
+    "\n",
+    "# if spoof < score, then bonafide, else spoof\n",
+    "pred_df['pred'] = pred_df.apply(\n",
+    "    lambda x: 'bonafide' if x['spoof'] < x['score'] else 'spoof', axis=1)\n",
+    "\n",
+    "# merge eval_df and pred_df on utt\n",
+    "res_df = pd.merge(full_df, pred_df, on='utt')\n",
+    "\n",
+    "print(\"Accuracy: {:.2f}\".format(\n",
+    "    accuracy_score(res_df[\"label\"], res_df[\"pred\"])*100))\n",
+    "print(\"\\n\")"
+   ]
   }
  ],
  "metadata": {
diff --git a/notebooks/prepare_dataset.ipynb b/notebooks/prepare_dataset.ipynb
index cd305f8..8617438 100644
--- a/notebooks/prepare_dataset.ipynb
+++ b/notebooks/prepare_dataset.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -10,7 +10,7 @@
      "output_type": "stream",
      "text": [
       "Reading protocol file...\n",
-      "Processing 408572 files using 8 workers...\n"
+      "Processing 248159 files using 8 workers...\n"
      ]
     },
     {
@@ -19,7 +19,7 @@
      "text": [
       "/home/hungdx/miniconda3/lib/python3.11/site-packages/numpy/core/fromnumeric.py:59: FutureWarning: 'DataFrame.swapaxes' is deprecated and will be removed in a future version. Please use 'DataFrame.transpose' instead.\n",
       "  return bound(*args, **kwds)\n",
-      "Calculating durations: 100%|██████████| 8/8 [00:49<00:00,  6.20s/it]\n"
+      "Calculating durations: 100%|██████████| 8/8 [00:02<00:00,  2.93it/s]\n"
      ]
     },
     {
@@ -30,10 +30,10 @@
       "Saving results...\n",
       "\n",
       "Processing Statistics:\n",
-      "Total files processed: 408572\n",
+      "Total files processed: 248159\n",
       "Failed files: 0\n",
-      "Total duration: 671.90 hours\n",
-      "Mean duration: 5.92 seconds\n",
+      "Total duration: 433.71 hours\n",
+      "Mean duration: 6.29 seconds\n",
       "Min duration: 0.08 seconds\n",
       "Max duration: 26.20 seconds\n",
       "\n",
@@ -148,8 +148,8 @@
     "\n",
     "# Usage\n",
     "BASE_DIR = \"/data/hungdx/Lightning-hydra/data/0_large-corpus\"\n",
-    "protocol_file = \"new_protocol_trim_vocoded.txt\"\n",
-    "output_file = \"audio_durations.csv\"\n",
+    "protocol_file = \"new_protocol_trim_vocoded_v2.txt\"\n",
+    "output_file = \"audio_durations_v2.csv\"\n",
     "\n",
     "# Run the processing\n",
     "df_results, stats = calculate_durations(\n",
@@ -166,7 +166,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -175,13 +175,13 @@
      "text": [
       "\n",
       "Short audio files (duration < 1 second):\n",
-      "2993\n"
+      "436\n"
      ]
     }
    ],
    "source": [
-    "# Filter audio files with duration less than 1 second\n",
-    "short_files = df_results[df_results['duration'] < 1]\n",
+    "# Filter audio files with duration less than 1 second with subset == train\n",
+    "short_files = df_results[df_results['duration'] < 1 & (df_results['subset'] == 'train')]\n",
     "print(\"\\nShort audio files (duration < 1 second):\")\n",
     "print(len(short_files))"
    ]
@@ -195,14 +195,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
-    "df_results = df_results[df_results['duration'] >= 1]\n",
+    "# remove all files with duration < 1 second and subset == train\n",
+    "df_results = df_results[~((df_results['duration'] < 1) & (df_results['subset'] == 'train'))]\n",
+    "\n",
     "# drop duration column and save\n",
     "df_results.drop(columns=['duration'], inplace=True)\n",
-    "df_results.to_csv(\"new_protocol_trim_vocoded_cleaned.txt\", index=False, header=False, sep=\" \")"
+    "df_results.to_csv(\"new_protocol_trim_vocoded_cleaned_v2.txt\", index=False, header=False, sep=\" \")"
    ]
   },
   {
@@ -231,6 +233,159 @@
     "\n",
     "print(\"After removing vocoded files:\", len(df))"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# MLAAD"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "df = pd.read_csv(\"~/MLAAD/protocol.txt\", sep=\" \", header=None)\n",
+    "df.columns = [\"utt_id\", \"subset\", \"unk\", \"label\"]\n",
+    "\n",
+    "# drop unk column\n",
+    "df.drop(columns=['unk'], inplace=True)\n",
+    "\n",
+    "df.to_csv(\"~/MLAAD/protocol_lts.txt\", index=False, header=False, sep=\" \")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Before removing vocoded files: 405579\n",
+      "After removing vocoded files: 390512\n"
+     ]
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "df = pd.read_csv(\"new_protocol_trim_vocoded_cleaned.txt\", sep=\" \", header=None)\n",
+    "df.columns = [\"utt_id\", \"subset\", \"label\"]\n",
+    "\n",
+    "print(\"Before removing vocoded files:\", len(df))\n",
+    "\n",
+    "df = df[~((df[\"utt_id\"].str.startswith(\"vocoded\")) & (df[\"subset\"] == \"dev\"))]\n",
+    "\n",
+    "print(\"After removing vocoded files:\", len(df))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# MLAAD V5"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Created protocol.txt with 152388 entries\n",
+      "Created combined_meta.csv with 152388 entries\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "import pandas as pd\n",
+    "from pathlib import Path\n",
+    "import csv\n",
+    "\n",
+    "\n",
+    "def read_csv_safely(file_path):\n",
+    "    try:\n",
+    "        # First attempt with standard reading\n",
+    "        return pd.read_csv(file_path, sep='|')\n",
+    "    except:\n",
+    "        try:\n",
+    "            # Second attempt with quote character handling\n",
+    "            return pd.read_csv(file_path, sep='|', quoting=csv.QUOTE_NONE, escapechar='\\\\')\n",
+    "        except:\n",
+    "            try:\n",
+    "                # Third attempt with error handling\n",
+    "                return pd.read_csv(file_path, sep='|', on_bad_lines='skip', quoting=csv.QUOTE_NONE)\n",
+    "            except Exception as e:\n",
+    "                print(f\"Error reading file {file_path}: {str(e)}\")\n",
+    "                return None\n",
+    "\n",
+    "\n",
+    "def create_protocol_and_merge_meta(root_dir):\n",
+    "    # Initialize empty list to store all metadata\n",
+    "    all_meta_data = []\n",
+    "\n",
+    "    # Initialize protocol file\n",
+    "    protocol_lines = []\n",
+    "\n",
+    "    # Walk through the directory structure\n",
+    "    for language_dir in os.listdir(root_dir):\n",
+    "        language_path = os.path.join(root_dir, language_dir)\n",
+    "        if not os.path.isdir(language_path):\n",
+    "            continue\n",
+    "\n",
+    "        # For each model directory in the language directory\n",
+    "        for model_dir in os.listdir(language_path):\n",
+    "            model_path = os.path.join(language_path, model_dir)\n",
+    "            if not os.path.isdir(model_path):\n",
+    "                continue\n",
+    "\n",
+    "            # Look for meta.csv file\n",
+    "            meta_file = os.path.join(model_path, 'meta.csv')\n",
+    "            if os.path.exists(meta_file):\n",
+    "                # Read meta.csv with safe reading function\n",
+    "                df = read_csv_safely(meta_file)\n",
+    "\n",
+    "                if df is not None:\n",
+    "                    # Add to combined metadata\n",
+    "                    all_meta_data.append(df)\n",
+    "\n",
+    "                    # Create protocol lines for all audio files in this directory\n",
+    "                    for audio_path in df['path']:\n",
+    "                        protocol_lines.append(f\"{audio_path} eval spoof\")\n",
+    "\n",
+    "    # Combine all metadata\n",
+    "    if all_meta_data:\n",
+    "        combined_meta = pd.concat(all_meta_data, ignore_index=True)\n",
+    "\n",
+    "        # Save combined metadata\n",
+    "        combined_meta.to_csv('mlaad_v5_combined_meta.csv', sep='|',\n",
+    "                             index=False, quoting=csv.QUOTE_NONE, escapechar='\\\\')\n",
+    "\n",
+    "        # Save protocol file\n",
+    "        with open('mlaad_v5_protocol.txt', 'w') as f:\n",
+    "            f.write('\\n'.join(protocol_lines))\n",
+    "\n",
+    "        print(f\"Created protocol.txt with {len(protocol_lines)} entries\")\n",
+    "        print(f\"Created combined_meta.csv with {len(combined_meta)} entries\")\n",
+    "    else:\n",
+    "        print(\"No meta.csv files found or all files had errors\")\n",
+    "\n",
+    "\n",
+    "# Run the function\n",
+    "if __name__ == \"__main__\":\n",
+    "    create_protocol_and_merge_meta(\n",
+    "        '/nvme1/hungdx/Lightning-hydra/data/mlaad_v5/fake')"
+   ]
   }
  ],
  "metadata": {
diff --git a/src/data/components/augwrapper.py b/src/data/components/augwrapper.py
index e959302..667a64d 100644
--- a/src/data/components/augwrapper.py
+++ b/src/data/components/augwrapper.py
@@ -14,7 +14,7 @@
 
 SUPPORTED_AUGMENTATION = [
     'background_noise_5_15', 'pitch_1', 'volume_10', 'reverb_1', 'speed_01', 'telephone_g722', 'gaussian_1', 'gaussian_2', 'gaussian_2_5', 'gaussian_3',
-    'RawBoostdf', 'RawBoost12', 'copy_paste_80', 'copy_paste_r', 'time_masking', 'masking', 'time_swap',
+    'RawBoostdf', 'RawBoost12', 'RawBoostFull', 'copy_paste_80', 'copy_paste_r', 'time_masking', 'masking', 'time_swap',
     'freq_swap', 'swapping', 'frequency_masking', 'linear_filter', 'mp32flac', 'ogg2flac', 'nonspeechtrim',
     'bandpass_0_4000', 'griffinlim_downsample', 'lowpass_hifigan_asvspoof5', 'lowpass_hifigan', 'librosa_downsample', 'none']
 
@@ -1054,6 +1054,24 @@ def RawBoost12(x, args, sr=16000, audio_path=None):
             return waveform
 
 
+def RawBoostFull(x, args, sr=16000, audio_path=None):
+    aug_dir = args.aug_dir
+    utt_id = os.path.basename(audio_path).split('.')[0]
+    aug_audio_path = os.path.join(aug_dir, 'RawBoostFull', utt_id + '.wav')
+    if args.online_aug:
+        return process_Rawboost_feature(x, sr, args, algo=4)
+    else:
+        # check if the augmented file exists
+        if (os.path.exists(aug_audio_path)):
+            waveform, _ = librosa.load(aug_audio_path, sr=sr, mono=True)
+            return waveform
+        else:
+            waveform = process_Rawboost_feature(x, sr, args, algo=5)
+            # save the augmented file,waveform in np array
+            sf.write(aug_audio_path, waveform, sr, subtype='PCM_16')
+            return waveform
+
+
 def RawBoostdf(x, args, sr=16000, audio_path=None):
     aug_dir = args.aug_dir
     utt_id = os.path.basename(audio_path).split('.')[0]
diff --git a/src/data/normal_multiview_datamodule.py b/src/data/normal_multiview_datamodule.py
index 093e41d..64d5c37 100644
--- a/src/data/normal_multiview_datamodule.py
+++ b/src/data/normal_multiview_datamodule.py
@@ -89,10 +89,10 @@ def __init__(self, args, list_IDs, labels, base_dir, algo=5, vocoders=[],
                                                trim_length, wav_samp_rate, noise_path, rir_path,
                                                aug_dir, online_aug, repeat_pad, is_train, random_start)
         self.enable_chunking = enable_chunking
-        if repeat_pad:
-            self.padding_type = "repeat"
-        else:
-            self.padding_type = "zero"
+        self.padding_type = "repeat" if repeat_pad else "zero"
+
+        print("trim_length:", trim_length)
+        print("padding_type:", self.padding_type)
 
     def __getitem__(self, idx):
         utt_id = self.list_IDs[idx]
@@ -263,7 +263,7 @@ def setup(self, stage: Optional[str] = None) -> None:
                                             base_dir=self.data_dir+'/',  is_train=False, **self.args['data'])
 
             self.data_test = Dataset_for_eval(self.args, list_IDs=file_eval, labels=None,
-                                              base_dir=self.data_dir+'/',  **self.args['data'])
+                                              base_dir=self.data_dir+'/',  random_start=self.args.random_start, trim_length=self.args.trim_length, repeat_pad=True if self.args.padding_type == 'repeat' else False)
 
     def train_dataloader(self) -> DataLoader[Any]:
         """Create and return the train dataloader.
@@ -367,9 +367,12 @@ def genList(self, is_train=False, is_eval=False, is_dev=False):
                 l_meta = f.readlines()
             for line in l_meta:
                 utt, subset, label = line.strip().split()
-                if subset == 'eval':
-                    file_list.append(utt)
-                    d_meta[utt] = 1 if label == 'bonafide' else 0
+                # if subset == 'eval':
+                #     file_list.append(utt)
+                #     d_meta[utt] = 1 if label == 'bonafide' else 0
+
+                file_list.append(utt)
+                d_meta[utt] = 1 if label == 'bonafide' else 0
             # return d_meta, file_list
             return d_meta, file_list