Update notebook

gemelo-ai · Jun 12, 2023 · 4220017 · 4220017
1 parent d768198
commit 4220017
Showing 1 changed file with 4 additions and 5 deletions.
diff --git a/notebooks/Bark+Vocos.ipynb b/notebooks/Bark+Vocos.ipynb
@@ -6,7 +6,7 @@
       "private_outputs": true,
       "provenance": [],
       "gpuType": "T4",
-      "authorship_tag": "ABX9TyNuxsqp/FTsmltYeYfMZ6sw",
+      "authorship_tag": "ABX9TyMC53IsYoVJIVijVzw3ADvX",
       "include_colab_link": true
     },
     "kernelspec": {
@@ -26,7 +26,7 @@
         "colab_type": "text"
       },
       "source": [
-        "<a href=\"https://colab.research.google.com/github/charactr-platform/vocos/blob/bark/notebooks/Bark%2BVocos.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
+        "<a href=\"https://colab.research.google.com/github/charactr-platform/vocos/blob/main/notebooks/Bark%2BVocos.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
       ]
     },
     {
@@ -41,7 +41,7 @@
     {
       "cell_type": "markdown",
       "source": [
-        "In this notebook, we use Bark generative model to turn a text prompt into EnCodec audio tokens. These tokens then go through two decoders, EnCodec and Vocos, to reconstruct the audio waveform. Compare the results to discover the differences in audio quality and characteristics."
+        "In this notebook, we use [Bark](https://github.com/suno-ai/bark) generative model to turn a text prompt into EnCodec audio tokens. These tokens then go through two decoders, EnCodec and Vocos, to reconstruct the audio waveform. Compare the results to discover the differences in audio quality and characteristics."
       ],
       "metadata": {
         "id": "zJFDte0daDAz"
@@ -103,6 +103,7 @@
       "cell_type": "code",
       "source": [
         "from vocos import Vocos\n",
+        "import torch\n",
         "\n",
         "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
         "vocos = Vocos.from_pretrained(\"charactr/vocos-encodec-24khz\").to(device)"
@@ -225,8 +226,6 @@
     {
       "cell_type": "code",
       "source": [
-        "import torch\n",
-        "\n",
         "audio_tokens_torch = torch.from_numpy(audio_tokens).to(device)\n",
         "features = vocos.codes_to_features(audio_tokens_torch)\n",
         "vocos_output = vocos.decode(features, bandwidth_id=torch.tensor([2], device=device))  # 6 kbps\n",