bark-with-voice-clone/generate.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bark.api import generate_audio\n",
    "from transformers import BertTokenizer\n",
    "from bark.generation import SAMPLE_RATE, preload_models, codec_decode, generate_coarse, generate_fine, generate_text_semantic\n",
    "\n",
    "# Enter your prompt and speaker here\n",
    "text_prompt = \"Hello, my name is Serpy. And, uh — and I like pizza. [laughs]\"\n",
    "voice_name = \"speaker_0\" # use your custom voice name here if you have one\n",
    "\n",
    "# load the tokenizer\n",
    "tokenizer = BertTokenizer.from_pretrained(\"bert-base-multilingual-cased\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# download and load all models\n",
    "preload_models(\n",
    "    text_use_gpu=True,\n",
    "    text_use_small=False,\n",
    "    coarse_use_gpu=True,\n",
    "    coarse_use_small=False,\n",
    "    fine_use_gpu=True,\n",
    "    fine_use_small=False,\n",
    "    codec_use_gpu=True,\n",
    "    force_reload=False,\n",
    "    path=\"models\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# simple generation\n",
    "audio_array = generate_audio(text_prompt, history_prompt=voice_name, text_temp=0.7, waveform_temp=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# generation with more control\n",
    "x_semantic = generate_text_semantic(\n",
    "    text_prompt,\n",
    "    history_prompt=voice_name,\n",
    "    temp=0.7,\n",
    "    top_k=50,\n",
    "    top_p=0.95,\n",
    ")\n",
    "\n",
    "x_coarse_gen = generate_coarse(\n",
    "    x_semantic,\n",
    "    history_prompt=voice_name,\n",
    "    temp=0.7,\n",
    "    top_k=50,\n",
    "    top_p=0.95,\n",
    ")\n",
    "x_fine_gen = generate_fine(\n",
    "    x_coarse_gen,\n",
    "    history_prompt=voice_name,\n",
    "    temp=0.5,\n",
    ")\n",
    "audio_array = codec_decode(x_fine_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import Audio\n",
    "# play audio\n",
    "Audio(audio_array, rate=SAMPLE_RATE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.io.wavfile import write as write_wav\n",
    "# save audio\n",
    "filepath = \"/output/audio.wav\" # change this to your desired output path\n",
    "write_wav(filepath, SAMPLE_RATE, audio_array)"
   ]
  },
  {
   "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": "3.10.8"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}