{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "3abd2681", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T03:55:47.955946Z", "start_time": "2023-08-22T03:55:44.515977Z" } }, "outputs": [], "source": [ "import os\n", "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"2\"\n", "\n", "from suno_utils.audio import Audio\n", "from suno_utils.tasks.mert_v2 import (\n", " encode as mert_encode,\n", " preload_models as preload_mert_models\n", ")\n", "from suno_utils.tasks.dac import (\n", " encode as codec_encode,\n", " decode as codec_decode,\n", " preload_models as preload_codec_models,\n", ")\n", "\n", "from suno_utils.tasks.gpt_v2.chirp_v1 import preload_models as preload_chirp_models\n", "from suno_utils.tasks.gpt_v2.chirp_v1 import generate_audio\n", "from suno_utils.tasks.gpt_v2.generation import generate\n", "\n", "from tqdm import tqdm" ] }, { "cell_type": "code", "execution_count": 2, "id": "f97f7e06", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T03:56:08.670283Z", "start_time": "2023-08-22T03:55:47.958522Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "model loaded: 1720.9M params, 1.52 loss\n", "WARNING: feature_extractor_cqt requires the libray 'nnAudio'\n" ] } ], "source": [ "preload_chirp_models(\n", " centroids_filepath=\"/home/mikeys/bundle/2x1k_centroids_mert.npy\",\n", " gpt_ckpt_path=\"/home/mikeys/bundle/xl_s5.pt\",\n", " codec_ckpt_path=\"/home/mikeys/bundle/d_codec_25x8.pt\",\n", " common_genre_tags_path=\"/home/mikeys/bundle/common_tags.json\",\n", ")" ] }, { "cell_type": "code", "execution_count": 3, "id": "cceb2397", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T03:56:08.674356Z", "start_time": "2023-08-22T03:56:08.671792Z" } }, "outputs": [], "source": [ "text = \"\"\"\n", "[Verse]\n", "In my kitchen, there's a problem I can't tame\n", "The sink is stuck, it's driving me insane\n", "Every dish piled up, I can't wash them away\n", "This sticky situation ruins my day\n", "\n", "[Chorus]\n", "Oh, kitchen sink, won't you set me free\n", "I despise this clog, it's my enemy\n", "Won't someone save me from this misery\n", "Oh, kitchen sink, release my agony\n", "\n", "[interlude]\n", "-\n", "\"\"\"" ] }, { "cell_type": "code", "execution_count": 4, "id": "7468dd06", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T04:16:47.408438Z", "start_time": "2023-08-22T03:56:08.675433Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "\r", " 0%| | 0/9 [00:00\n", " \n", " Your browser does not support the audio element.\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "50\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "100\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "500\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "750\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "1000\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "1500\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "2000\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "5000\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "for k, v in temp_lookup.items():\n", " print(k)\n", " for o in v:\n", " o.play()" ] }, { "cell_type": "code", "execution_count": 6, "id": "f1cbb567", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T04:16:53.592352Z", "start_time": "2023-08-22T04:16:52.969405Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1\n", "50\n", "100\n", "500\n", "750\n", "1000\n", "1500\n", "2000\n", "5000\n" ] } ], "source": [ "for k, v in temp_lookup.items():\n", " print(k)\n", " for i, o in enumerate(v):\n", " o.write_wav(f\"audios/kitchen_sink_top_k_coarse_{k}_{i}.wav\")" ] }, { "cell_type": "code", "execution_count": 7, "id": "49eed3b9", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T04:16:53.595283Z", "start_time": "2023-08-22T04:16:53.593602Z" } }, "outputs": [], "source": [ "# import json\n", "# with open(\"/home/mikeys/bundle/common_tags.json\", \"r\") as fp:\n", "# common_tags = json.load(fp)" ] }, { "cell_type": "code", "execution_count": 8, "id": "37d9f742", "metadata": { "ExecuteTime": { "end_time": "2023-08-22T04:16:53.642743Z", "start_time": "2023-08-22T04:16:53.596412Z" } }, "outputs": [], "source": [ "# common_tags" ] }, { "cell_type": "code", "execution_count": null, "id": "e3c7d909", "metadata": {}, "outputs": [], "source": [] } ], "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": "3.10.12" }, "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": 5 }