sentiment_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-15 03:11:24.579891: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-10-15 03:11:24.637735: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-10-15 03:11:24.638555: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-10-15 03:11:25.801037: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import scipy.io.wavfile as wav\n",
    "from speechpy.feature import mfcc\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_signal_length = 88500"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_feature_vector_from_mfcc(file_path: str, flatten: bool, mfcc_len: int = 39) -> np.ndarray:\n",
    "    fs, signal = wav.read(file_path)\n",
    "    s_len = len(signal)\n",
    "    # pad the signals to have same size if lesser than required\n",
    "    # else slice them\n",
    "    if s_len < mean_signal_length:\n",
    "        pad_len = mean_signal_length - s_len\n",
    "        pad_rem = pad_len % 2\n",
    "        pad_len //= 2\n",
    "        signal = np.pad(signal, (pad_len, pad_len + pad_rem),\n",
    "                        'constant', constant_values=0)\n",
    "    else:\n",
    "        pad_len = s_len - mean_signal_length\n",
    "        pad_len //= 2\n",
    "        signal = signal[pad_len:pad_len + mean_signal_length]\n",
    "    mel_coefficients = mfcc(signal, fs, num_cepstral=mfcc_len)\n",
    "    if flatten:\n",
    "        # Flatten the data\n",
    "        mel_coefficients = np.ravel(mel_coefficients)\n",
    "    return mel_coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = './voice_database/test2/1.wav'\n",
    "features = get_feature_vector_from_mfcc(filename, flatten=False)\n",
    "features = np.expand_dims(features, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = [\"Neutral\", \"Angry\", \"Happy\", \"Sad\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-15 03:17:59.451229: E tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:268] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n"
     ]
    }
   ],
   "source": [
    "model = tf.keras.models.load_model('./sentiment_models/sentiment_classifier.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 19ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'Sad'"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = np.argmax(model.predict(np.array([features])))\n",
    "pred_label = labels[pred]\n",
    "pred_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "voice",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}