{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'list'>\n",
      "(1, 478720)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np, json, yaml\n",
    "from scipy.io.wavfile import write\n",
    "\n",
    "with open('/projects/mixart/doodle-musicgen/.conf/generation_conf.yaml') as f:\n",
    "  conf = yaml.safe_load(f)\n",
    "with open('response.json','r') as f:\n",
    "  res = json.load(f)\n",
    "audio = res['audio']\n",
    "print(type(audio))\n",
    "audio = np.array(audio)\n",
    "print(audio.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "rate = conf['sampling_rate']\n",
    "scaled = np.int16(audio.flatten() * int(2**16//2-1))\n",
    "write('response.wav', rate, scaled)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ezout-vision",
   "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": "3.10.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}