{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:24.487521Z",
     "start_time": "2020-12-14T13:43:23.780570Z"
    }
   },
   "outputs": [],
   "source": [
    "import glob\n",
    "import torch\n",
    "from IPython.display import Audio\n",
    "torch.set_num_threads(1)\n",
    "from utils import (init_jit_model, get_speech_ts, \n",
    "                  save_audio, read_audio, \n",
    "                  state_generator, single_audio_stream, init_onnx_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Full audio example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:24.492506Z",
     "start_time": "2020-12-14T13:43:24.489440Z"
    }
   },
   "outputs": [],
   "source": [
    "def collect_speeches(tss, wav):\n",
    "    speech_chunks = []\n",
    "    for i in tss:\n",
    "        speech_chunks.append(wav[i['start']: i['end']])\n",
    "    return torch.cat(speech_chunks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:24.760714Z",
     "start_time": "2020-12-14T13:43:24.493992Z"
    }
   },
   "outputs": [],
   "source": [
    "model = init_jit_model('files/joint_VAD_just_RU_jit_cut_q.pth.tar', 'cpu')  # from yml file\n",
    "model = init_onnx_model('files/joint_VAD_just_RU.onnx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:24.793384Z",
     "start_time": "2020-12-14T13:43:24.762311Z"
    }
   },
   "outputs": [],
   "source": [
    "Audio('files/test_audio_2.wav')\n",
    "wav = read_audio('files/test_audio_2.wav')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:25.320324Z",
     "start_time": "2020-12-14T13:43:24.808594Z"
    }
   },
   "outputs": [],
   "source": [
    "speech_timestamps = get_speech_ts(wav, model, num_steps=4)  # kill extractor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:25.324901Z",
     "start_time": "2020-12-14T13:43:25.321759Z"
    }
   },
   "outputs": [],
   "source": [
    "speech_timestamps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:25.344065Z",
     "start_time": "2020-12-14T13:43:25.326162Z"
    }
   },
   "outputs": [],
   "source": [
    "save_audio('only_speech.wav', collect_speeches(speech_timestamps, wav), 16000)\n",
    "Audio('only_speech.wav')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Single stream example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:25.778585Z",
     "start_time": "2020-12-14T13:43:25.496583Z"
    }
   },
   "outputs": [],
   "source": [
    "model = init_jit_model('files/joint_VAD_just_RU_jit_cut_q.pth.tar', 'cpu')  # from yml file\n",
    "#model = init_onnx_model('files/joint_VAD_just_RU.onnx')\n",
    "audio = 'files/test_audio_6.wav'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:29.402604Z",
     "start_time": "2020-12-14T13:43:25.780037Z"
    }
   },
   "outputs": [],
   "source": [
    "for i in single_audio_stream(model, audio):\n",
    "    if i:\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multiple stream example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:29.674262Z",
     "start_time": "2020-12-14T13:43:29.403972Z"
    }
   },
   "outputs": [],
   "source": [
    "model = init_jit_model('files/joint_VAD_just_RU_jit_cut_q.pth.tar', 'cpu')  # from yml file\n",
    "model = init_onnx_model('files/joint_VAD_just_RU.onnx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:29.678449Z",
     "start_time": "2020-12-14T13:43:29.675519Z"
    }
   },
   "outputs": [],
   "source": [
    "audios_for_stream = glob.glob('files/test*.wav')\n",
    "len(audios_for_stream)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:43:40.236387Z",
     "start_time": "2020-12-14T13:43:29.679274Z"
    }
   },
   "outputs": [],
   "source": [
    "for i in state_generator(model, audios_for_stream, audios_in_stream=2):\n",
    "    if i:\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-12-14T13:46:49.812052Z",
     "start_time": "2020-12-14T13:46:49.586637Z"
    }
   },
   "outputs": [],
   "source": [
    "!cp ../silero-models-research/model_saves/joint_VAD_just_RU_jit_cut_q.pth.tar files/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}