{ "cells": [ { "attachments": {}, "cell_type": "markdown", "id": "285a34a4-18ff-493c-8322-71adf1f09622", "metadata": {}, "source": [ "# Audio compression with EnCodec and OpenVINO\n", "\n", "Compression is an important part of the Internet today because it enables people to easily share high-quality photos, listen to audio messages, stream their favorite shows, and so much more. Even when using today’s state-of-the-art techniques, enjoying these rich multimedia experiences requires a high speed Internet connection and plenty of storage space. AI helps to overcome these limitations: \"Imagine listening to a friend’s audio message in an area with low connectivity and not having it stall or glitch.\"\n", "\n", "This tutorial considers ways to use OpenVINO and EnCodec algorithm for hyper compression of audio.\n", "EnCodec is a real-time, high-fidelity audio codec that uses AI to compress audio files without losing quality. It was introduced in [High Fidelity Neural Audio Compression](https://arxiv.org/pdf/2210.13438.pdf) paper by Meta AI. The researchers claimed they achieved an approximate 10x compression rate without loss of quality and made it work for CD-quality audio. More details about this approach can be found in [Meta AI blog](https://ai.facebook.com/blog/ai-powered-audio-compression-technique/) and original [repo](https://github.com/facebookresearch/encodec).\n", "\n", "![image.png](https://github.com/openvinotoolkit/openvino_notebooks/assets/29454499/17546d66-12b9-4841-9293-cc878258a186)\n", "\n", "\n", "#### Table of contents:\n", "\n", "- [Prerequisites](#Prerequisites)\n", "- [Instantiate audio compression pipeline](#Instantiate-audio-compression-pipeline)\n", "- [Explore EnCodec pipeline](#Explore-EnCodec-pipeline)\n", " - [Preprocessing](#Preprocessing)\n", " - [Encoding](#Encoding)\n", " - [Decompression](#Decompression)\n", "- [Convert model to OpenVINO Intermediate Representation format](#Convert-model-to-OpenVINO-Intermediate-Representation-format)\n", "- [Integrate OpenVINO to EnCodec pipeline](#Integrate-OpenVINO-to-EnCodec-pipeline)\n", " - [Select inference device](#Select-inference-device)\n", "- [Run EnCodec with OpenVINO](#Run-EnCodec-with-OpenVINO)\n", "\n" ] }, { "attachments": {}, "cell_type": "markdown", "id": "e328a76f-03b8-4972-afa9-36d3f8cc8364", "metadata": {}, "source": [ "## Prerequisites\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "Install required dependencies:" ] }, { "cell_type": "code", "execution_count": 1, "id": "c2ec9aa3-00c2-4d49-954c-c1af2fe43f2a", "metadata": {}, "outputs": [], "source": [ "%pip install -q --extra-index-url https://download.pytorch.org/whl/cpu \"openvino>=2023.3.0\" \"torch>=2.1\" \"torchaudio>=2.1\" \"encodec>=0.1.1\" \"gradio>=4.19\" \"librosa>=0.8.1\" \"matplotlib<=3.7\" tqdm" ] }, { "attachments": {}, "cell_type": "markdown", "id": "82279f60-a018-4ea8-9b0b-77f0f40903cf", "metadata": {}, "source": [ "## Instantiate audio compression pipeline\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "[Codecs](https://en.wikipedia.org/wiki/Codec), which act as encoders and decoders for streams of data, help empower most of the audio compression people currently use online. Some examples of commonly used codecs include MP3, Opus, and EVS. Classic codecs like these decompose the signal between different frequencies and encode as efficiently as possible. Most classic codecs leverage human hearing knowledge (psychoacoustics) but have a finite or given set of handcrafted ways to efficiently encode and decode the file. EnCodec, a neural network that is trained from end to end to reconstruct the input signal, was introduced as an attempt to overcome this limitation. It consists of three parts:\n", "\n", "* The **encoder**, which takes the uncompressed data in and transforms it into a higher dimensional and lower frame rate representation.\n", "\n", "* The **quantizer**, which compresses this representation to the target size. This compressed representation is what is stored on disk or will be sent through the network.\n", "\n", "* The **decoder** is the final step. It turns the compressed signal back into a waveform that is as similar as possible to the original. The key to lossless compression is to identify changes that will not be perceivable by humans, as perfect reconstruction is impossible at low bit rates.\n", "\n", "![encodec_compression](https://github.com/openvinotoolkit/openvino_notebooks/assets/29454499/5cd9a482-b42b-4dea-85a5-6d66b20ce13d))\n", "\n", "\n", "The authors provide two multi-bandwidth models:\n", "* `encodec_model_24khz` - a causal model operating at 24 kHz on monophonic audio trained on a variety of audio data.\n", "* `encodec_model_48khz` - a non-causal model operating at 48 kHz on stereophonic audio trained on music-only data.\n", "\n", "In this tutorial, we will use `encodec_model_24khz` as an example, but the same actions are also applicable to `encodec_model_48khz` model as well.\n", "To start working with this model, we need to instantiate model class using `EncodecModel.encodec_model_24khz()` and select required compression bandwidth among available: 1.5, 3, 6, 12 or 24 kbps for 24 kHz model and 3, 6, 12 and 24 kbps for 48 kHz model. We will use 6 kbps bandwidth." ] }, { "cell_type": "code", "execution_count": 2, "id": "d9855472-5c0d-4853-b284-ed659885c76e", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/ea/work/genai_env/lib/python3.8/site-packages/torch/nn/utils/weight_norm.py:30: UserWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\n", " warnings.warn(\"torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\")\n" ] } ], "source": [ "from encodec import compress, decompress\n", "from encodec.utils import convert_audio, save_audio\n", "from encodec.compress import MODELS\n", "import torchaudio\n", "import torch\n", "import typing as tp\n", "\n", "model_id = \"encodec_24khz\"\n", "# Instantiate a pretrained EnCodec model\n", "model = MODELS[model_id]()\n", "model.set_target_bandwidth(6.0)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "2146fbd2-6f15-4e31-b9f6-7c8e69daabd0", "metadata": {}, "source": [ "## Explore EnCodec pipeline\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "Let us explore model capabilities on example audio:" ] }, { "cell_type": "code", "execution_count": 3, "id": "84f727cf-9414-480a-b824-cae3bb97371e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "'test_24k.wav' already exists.\n" ] }, { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import librosa\n", "import matplotlib.pyplot as plt\n", "import librosa.display\n", "import IPython.display as ipd\n", "\n", "# Fetch `notebook_utils` module\n", "import requests\n", "\n", "r = requests.get(\n", " url=\"https://raw.githubusercontent.com/openvinotoolkit/openvino_notebooks/latest/utils/notebook_utils.py\",\n", ")\n", "\n", "open(\"notebook_utils.py\", \"w\").write(r.text)\n", "from notebook_utils import download_file\n", "\n", "test_data_url = \"https://github.com/facebookresearch/encodec/raw/main/test_24k.wav\"\n", "\n", "sample_file = \"test_24k.wav\"\n", "download_file(test_data_url, sample_file)\n", "audio, sr = librosa.load(sample_file)\n", "plt.figure(figsize=(14, 5))\n", "librosa.display.waveshow(audio, sr=sr)\n", "\n", "ipd.Audio(sample_file)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "970e162d-ea7e-4d32-bb35-3c3ed29f9720", "metadata": {}, "source": [ "### Preprocessing\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "To achieve the best result, audio should have the number of channels and sample rate expected by the model. If audio does not fulfill these requirements, it can be converted to the desired sample rate and the number of channels using the `convert_audio` function." ] }, { "cell_type": "code", "execution_count": 4, "id": "e2704048-da45-46ca-bd76-2083194c9208", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model expected sample rate 24000\n", "Model expected audio format mono\n" ] } ], "source": [ "model_sr, model_channels = model.sample_rate, model.channels\n", "print(f\"Model expected sample rate {model_sr}\")\n", "print(f\"Model expected audio format {'mono' if model_channels == 1 else 'stereo'}\")" ] }, { "cell_type": "code", "execution_count": 5, "id": "98235cce-41d5-44dc-b543-85f6ec074a97", "metadata": {}, "outputs": [], "source": [ "# Load and pre-process the audio waveform\n", "wav, sr = torchaudio.load(sample_file)\n", "\n", "wav = convert_audio(wav, sr, model_sr, model_channels)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "f51bfa76-25ad-427c-851a-fb61481f02e3", "metadata": {}, "source": [ "### Encoding\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "Audio waveform should be split by chunks and then encoded by Encoder model, then compressed by quantizer for reducing memory. The result of compression is a binary file with `ecdc` extension, a special format for storing EnCodec compressed audio on disc." ] }, { "cell_type": "code", "execution_count": 6, "id": "0a137999-a1f8-4ac5-ac64-5afe38cb4b1c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "15067" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from pathlib import Path\n", "\n", "\n", "out_file = Path(\"compressed.ecdc\")\n", "b = compress(model, wav)\n", "out_file.write_bytes(b)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "22b28212-9cf1-4f89-ac92-da469239e87e", "metadata": {}, "source": [ "Let us compare obtained compression result:" ] }, { "cell_type": "code", "execution_count": 7, "id": "6cac0be4-3d01-4431-b56c-c4dfb67eb69d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "size before compression in Bytes: 960078\n", "size after compression in Bytes: 15067\n", "Compression file size ratio: 63.72\n" ] } ], "source": [ "import os\n", "\n", "orig_file_stats = os.stat(sample_file)\n", "compressed_file_stats = os.stat(\"compressed.ecdc\")\n", "print(f\"size before compression in Bytes: {orig_file_stats.st_size}\")\n", "print(f\"size after compression in Bytes: {compressed_file_stats.st_size}\")\n", "print(f\"Compression file size ratio: {orig_file_stats.st_size / compressed_file_stats.st_size:.2f}\")" ] }, { "attachments": {}, "cell_type": "markdown", "id": "94c1a9dd-09bc-4369-8b55-935db09c729b", "metadata": {}, "source": [ "Great! Now, we see the power of hyper compression. Binary size of a file becomes 60 times smaller and more suitable for sending via network." ] }, { "attachments": {}, "cell_type": "markdown", "id": "41301b36-2757-48c7-91dd-5c2d38d5c467", "metadata": {}, "source": [ "### Decompression\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "After successful sending of the compressed audio, it should be decompressed on the recipient's side. The decoder model is responsible for restoring the compressed signal back into a waveform that is as similar as possible to the original." ] }, { "cell_type": "code", "execution_count": 8, "id": "9cc5d08d-1a65-493e-b3fb-fd70dd2efbf7", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/ea/work/genai_env/lib/python3.8/site-packages/torch/nn/utils/weight_norm.py:30: UserWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\n", " warnings.warn(\"torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\")\n" ] } ], "source": [ "out, out_sr = decompress(out_file.read_bytes())" ] }, { "cell_type": "code", "execution_count": 9, "id": "196ca5a7-b9b0-4993-b1f0-3d81be1ce425", "metadata": {}, "outputs": [], "source": [ "output_file = \"decopressed.wav\"\n", "save_audio(out, output_file, out_sr)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "40e394db-512b-4b87-bfd0-fd5afe8ef1a3", "metadata": {}, "source": [ "The decompressed audio will be saved to the `decompressed.wav` file when decompression is finished. We can compare result with the original audio." ] }, { "cell_type": "code", "execution_count": 10, "id": "4ecafd48-f8db-48bb-a9d7-a41f00964c28", "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "audio, sr = librosa.load(output_file)\n", "plt.figure(figsize=(14, 5))\n", "librosa.display.waveshow(audio, sr=sr)\n", "\n", "ipd.Audio(output_file)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "ac4cc2e5-fd24-409e-8104-77d8c24bfa7d", "metadata": {}, "source": [ "Nice! Audio sounds close to original." ] }, { "attachments": {}, "cell_type": "markdown", "id": "44271e54-ea69-4da6-a84b-838b7896c586", "metadata": {}, "source": [ "## Convert model to OpenVINO Intermediate Representation format\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "For best results with OpenVINO, it is recommended to convert the model to OpenVINO IR format. OpenVINO supports PyTorch via conversion to OpenVINO IR format. We need to provide initialized model's instance and example of inputs for shape inference. We will use `ov.convert_model` functionality to convert the PyTorch models. The `ov.convert_model` Python function returns an OpenVINO model ready to load on the device and start making predictions. We can save it on disk for the next usage with `ov.save_model`." ] }, { "cell_type": "code", "execution_count": 11, "id": "7f37ce32-452e-4aa4-8311-16c7439da8e8", "metadata": {}, "outputs": [], "source": [ "class FrameEncoder(torch.nn.Module):\n", " def __init__(self, model):\n", " super().__init__()\n", " self.model = model\n", "\n", " def forward(self, x: torch.Tensor):\n", " codes, scale = self.model._encode_frame(x)\n", " if not self.model.normalize:\n", " return codes\n", " return codes, scale" ] }, { "cell_type": "code", "execution_count": 12, "id": "1948a427-7a5d-4520-a75d-ea4a0f11204f", "metadata": {}, "outputs": [], "source": [ "class FrameDecoder(torch.nn.Module):\n", " def __init__(self, model):\n", " super().__init__()\n", " self.model = model\n", "\n", " def forward(self, codes, scale=None):\n", " return model._decode_frame((codes, scale))" ] }, { "cell_type": "code", "execution_count": 13, "id": "07a7f491-c571-464f-a2be-7eb364e78325", "metadata": {}, "outputs": [], "source": [ "encoder = FrameEncoder(model)\n", "decoder = FrameDecoder(model)" ] }, { "cell_type": "code", "execution_count": 14, "id": "36df95d8-a254-4061-9a56-7aa4a03d11a4", "metadata": {}, "outputs": [], "source": [ "import openvino as ov\n", "\n", "\n", "core = ov.Core()\n", "\n", "OV_ENCODER_PATH = Path(\"encodec_encoder.xml\")\n", "if not OV_ENCODER_PATH.exists():\n", " encoder_ov = ov.convert_model(encoder, example_input=torch.zeros(1, 1, 480000), input=[[1, 1, 480000]])\n", " ov.save_model(encoder_ov, OV_ENCODER_PATH)\n", "else:\n", " encoder_ov = core.read_model(OV_ENCODER_PATH)" ] }, { "cell_type": "code", "execution_count": 15, "id": "480bb4e7-0958-46ec-b24d-7f4af77bc671", "metadata": {}, "outputs": [], "source": [ "OV_DECODER_PATH = Path(\"encodec_decoder.xml\")\n", "if not OV_DECODER_PATH.exists():\n", " decoder_ov = ov.convert_model(\n", " decoder,\n", " example_input=torch.zeros([1, 8, 1500], dtype=torch.long),\n", " input=[[1, 8, 1500]],\n", " )\n", " ov.save_model(decoder_ov, OV_DECODER_PATH)\n", "else:\n", " decoder_ov = core.read_model(OV_DECODER_PATH)" ] }, { "attachments": {}, "cell_type": "markdown", "id": "bc8cf59c-a3fa-48ea-9f18-0979860f0aa9", "metadata": {}, "source": [ "## Integrate OpenVINO to EnCodec pipeline\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "The following steps are required for integration of OpenVINO to EnCodec pipeline:\n", "\n", "1. Load the model to a device.\n", "2. Define audio frame processing functions.\n", "3. Replace the original frame processing functions with OpenVINO based algorithms." ] }, { "attachments": {}, "cell_type": "markdown", "id": "e32cea0b-e92c-40e7-912f-b5d47a43f921", "metadata": {}, "source": [ "### Select inference device\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "select device from dropdown list for running inference using OpenVINO" ] }, { "cell_type": "code", "execution_count": 16, "id": "43944e3c-4734-45ac-9c1f-2a70adc047e7", "metadata": {}, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "070c9129951e4c638d80a6364d3828c9", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Dropdown(description='Device:', index=3, options=('CPU', 'GPU.0', 'GPU.1', 'AUTO'), value='AUTO')" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import ipywidgets as widgets\n", "\n", "device = widgets.Dropdown(\n", " options=core.available_devices + [\"AUTO\"],\n", " value=\"AUTO\",\n", " description=\"Device:\",\n", " disabled=False,\n", ")\n", "\n", "device" ] }, { "cell_type": "code", "execution_count": 17, "id": "ba53f8f9-18b6-4f68-9f9e-a800c2f5b1a6", "metadata": {}, "outputs": [], "source": [ "compiled_encoder = core.compile_model(encoder_ov, device.value)\n", "encoder_out = compiled_encoder.output(0)\n", "\n", "compiled_decoder = core.compile_model(decoder_ov, device.value)\n", "decoder_out = compiled_decoder.output(0)" ] }, { "cell_type": "code", "execution_count": 18, "id": "967846a5-164b-4778-93b0-9b5947d93104", "metadata": {}, "outputs": [], "source": [ "def encode_frame(x: torch.Tensor):\n", " has_scale = len(compiled_encoder.outputs) == 2\n", " result = compiled_encoder(x)\n", " codes = torch.from_numpy(result[encoder_out])\n", " if has_scale:\n", " scale = torch.from_numpy(result[compiled_encoder.output(1)])\n", " else:\n", " scale = None\n", " return codes, scale" ] }, { "cell_type": "code", "execution_count": 19, "id": "e0886546-08bb-4b01-8bb3-c8ed7c722fd0", "metadata": {}, "outputs": [], "source": [ "EncodedFrame = tp.Tuple[torch.Tensor, tp.Optional[torch.Tensor]]\n", "\n", "\n", "def decode_frame(encoded_frame: EncodedFrame):\n", " codes, scale = encoded_frame\n", " inputs = [codes]\n", " if scale is not None:\n", " inputs.append(scale)\n", " return torch.from_numpy(compiled_decoder(inputs)[decoder_out])" ] }, { "cell_type": "code", "execution_count": 20, "id": "16681a9a-cd29-4ce6-838c-09f4514eed7a", "metadata": {}, "outputs": [], "source": [ "model._encode_frame = encode_frame\n", "model._decode_frame = decode_frame\n", "\n", "MODELS[model_id] = lambda: model" ] }, { "attachments": {}, "cell_type": "markdown", "id": "21b2e5b4-035c-4123-8257-00cd3f4af3fc", "metadata": {}, "source": [ "## Run EnCodec with OpenVINO\n", "[back to top ⬆️](#Table-of-contents:)\n", "\n", "The process of running encodec with OpenVINO under hood will be the same like with the original PyTorch models." ] }, { "cell_type": "code", "execution_count": 21, "id": "9213d6b6-56c1-4bbe-aa18-4bd1006513b0", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "15067" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "b = compress(model, wav, use_lm=False)\n", "out_file = Path(\"compressed_ov.ecdc\")\n", "out_file.write_bytes(b)" ] }, { "cell_type": "code", "execution_count": 22, "id": "ce575606-fe03-4996-8fe2-f64deda0e766", "metadata": {}, "outputs": [], "source": [ "out, out_sr = decompress(out_file.read_bytes())" ] }, { "cell_type": "code", "execution_count": 23, "id": "a297c13b-de70-4c96-85a0-d173667de241", "metadata": {}, "outputs": [], "source": [ "ov_output_file = \"decopressed_ov.wav\"\n", "save_audio(out, ov_output_file, out_sr)" ] }, { "cell_type": "code", "execution_count": 24, "id": "fd5d876a-648c-43b6-bbf3-def1f4cbb118", "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "audio, sr = librosa.load(ov_output_file)\n", "plt.figure(figsize=(14, 5))\n", "librosa.display.waveshow(audio, sr=sr)\n", "\n", "ipd.Audio(ov_output_file)" ] }, { "cell_type": "code", "execution_count": null, "id": "030a34e1-4b3d-4aec-aae1-633772e95992", "metadata": {}, "outputs": [], "source": [ "import gradio as gr\n", "from typing import Tuple\n", "import numpy as np\n", "\n", "\n", "def preprocess(input, sample_rate, model_sr, model_channels):\n", " input = torch.tensor(input, dtype=torch.float32)\n", " input = input / 2**15 # adjust to int16 scale\n", " input = input.unsqueeze(0)\n", " input = convert_audio(input, sample_rate, model_sr, model_channels)\n", "\n", " return input\n", "\n", "\n", "def postprocess(output):\n", " output = output.squeeze()\n", " output = output * 2**15 # adjust to [-1, 1] scale\n", " output = output.numpy(force=True)\n", " output = output.astype(np.int16)\n", "\n", " return output\n", "\n", "\n", "def _compress(input: Tuple[int, np.ndarray]):\n", " sample_rate, waveform = input\n", " waveform = preprocess(waveform, sample_rate, model_sr, model_channels)\n", "\n", " b = compress(model, waveform, use_lm=False)\n", " out, out_sr = decompress(b)\n", "\n", " out = postprocess(out)\n", " return out_sr, out\n", "\n", "\n", "demo = gr.Interface(_compress, \"audio\", \"audio\", examples=[\"test_24k.wav\"])\n", "\n", "try:\n", " demo.launch(debug=True)\n", "except Exception:\n", " demo.launch(share=True, debug=True)\n", "# if you are launching remotely, specify server_name and server_port\n", "# demo.launch(server_name='your server name', server_port='server port in int')\n", "# Read more in the docs: https://gradio.app/docs/" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.8.10" }, "openvino_notebooks": { "imageUrl": "https://github.com/openvinotoolkit/openvino_notebooks/blob/latest/notebooks/encodec-audio-compression/encodec-audio-compression.png?raw=true", "tags": { "categories": [ "Model Demos" ], "libraries": [], "other": [], "tasks": [ "Audio Compression" ] } }, "widgets": { "application/vnd.jupyter.widget-state+json": { "state": {}, "version_major": 2, "version_minor": 0 } } }, "nbformat": 4, "nbformat_minor": 5 }