basic-coquitts-usage-in-kaggle.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":7258506,"sourceType":"datasetVersion","datasetId":4206336}],"dockerImageVersionId":30627,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/yoonjunggyu/basic-coquitts-usage-in-kaggle?scriptVersionId=156501899\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown"},{"cell_type":"markdown","source":"1. Press \"Copy & Edit\".\n2. Press \"Run All\".\n3.\n\n![image](https://github.com/yjg30737/coquiTTS-kaggle/assets/55078043/febd5e89-f6a3-4c84-99de-1187d1780323)\n\nPress enter, and then everything will be a-OK.","metadata":{}},{"cell_type":"code","source":"!pip install TTS\n!pip install -U torch torchaudio","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:36:33.995004Z","iopub.execute_input":"2023-12-26T01:36:33.995488Z","iopub.status.idle":"2023-12-26T01:37:00.544333Z","shell.execute_reply.started":"2023-12-26T01:36:33.995455Z","shell.execute_reply":"2023-12-26T01:37:00.543282Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Import necessary packages","metadata":{}},{"cell_type":"code","source":"import os\nimport torch\nfrom TTS.api import TTS","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:37:00.547165Z","iopub.execute_input":"2023-12-26T01:37:00.547795Z","iopub.status.idle":"2023-12-26T01:37:00.552688Z","shell.execute_reply.started":"2023-12-26T01:37:00.547746Z","shell.execute_reply":"2023-12-26T01:37:00.551806Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Variables & Making output directory","metadata":{}},{"cell_type":"code","source":"default_path = 'output'\n# Get device\ndevice = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\nspeaker_wav = \"../input/voice-of-serana/serana.mp3\"\nos.makedirs(default_path, exist_ok=True)","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:37:00.55392Z","iopub.execute_input":"2023-12-26T01:37:00.554206Z","iopub.status.idle":"2023-12-26T01:37:00.563891Z","shell.execute_reply.started":"2023-12-26T01:37:00.554181Z","shell.execute_reply":"2023-12-26T01:37:00.562965Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def run(text, filename):\n    filename = os.path.join(default_path, f'{filename}.mp3')\n    tts.tts_to_file(text=text, speaker_wav=speaker_wav, language=\"en\", file_path=filename)","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:37:00.565016Z","iopub.execute_input":"2023-12-26T01:37:00.565344Z","iopub.status.idle":"2023-12-26T01:37:00.573163Z","shell.execute_reply.started":"2023-12-26T01:37:00.565313Z","shell.execute_reply":"2023-12-26T01:37:00.572279Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Load model","metadata":{}},{"cell_type":"code","source":"# Init TTS\ntts = TTS(\"tts_models/multilingual/multi-dataset/xtts_v2\").to(device)","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:37:00.575853Z","iopub.execute_input":"2023-12-26T01:37:00.576251Z","iopub.status.idle":"2023-12-26T01:43:16.550962Z","shell.execute_reply.started":"2023-12-26T01:37:00.576217Z","shell.execute_reply":"2023-12-26T01:43:16.550142Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Run","metadata":{}},{"cell_type":"code","source":"# Run TTS\n# ❗ Since this model is multi-lingual voice cloning model, we must set the target speaker_wav and language\n# Text to speech list of amplitude values as output\n# Text to speech to a file\n\ncontent = '''\nIf Janus, one of Saturn's moons, were to fall towards Earth, the outcome would depend on a variety of factors, including the moon's composition, speed, angle of entry into the Earth's atmosphere, and the Earth's gravitational influence.\nJanus is a small moon with a mean radius of about 89.5 kilometers. If it were to enter the Earth's atmosphere, its size would mean that it wouldn't burn up like a small meteorite. Instead, its entry would likely cause immense atmospheric disruption.\nThe most significant threat would be if Janus made it through the atmosphere and impacted the Earth. Given its size, an impact would be catastrophic, potentially causing global climatic changes and mass extinctions. The energy released would be several magnitudes greater than the largest nuclear detonations.\nEven before impact, the gravitational pull of such a large object would have severe effects on Earth. It could cause massive tidal waves, volcanic activity, and earthquakes.\nDepending on various factors like its internal structure and the angle of entry, Janus could fragment before impact. However, even fragmented, the pieces would still cause widespread devastation.\nWhile a small part of its mass might burn due to friction with the Earth's atmosphere, this would be insignificant compared to the overall mass of Janus.\nRealistically, for Janus to fall towards Earth, it would require an enormous alteration in its current orbit around Saturn, which is highly improbable under normal celestial mechanics.\nIn summary, if Janus were to fall to Earth, it wouldn't simply burn up in the atmosphere. The consequences would be globally catastrophic, potentially leading to mass extinctions and significant alterations to the Earth's surface and climate.\n'''\n\n# run(content, 'output.mp3')\n\nc_arr = list(filter(lambda x: x, content.split('\\n')))\nfor i, c in enumerate(c_arr):\n    run(c, i)","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-12-26T01:43:16.552316Z","iopub.execute_input":"2023-12-26T01:43:16.553275Z","iopub.status.idle":"2023-12-26T01:53:03.424839Z","shell.execute_reply.started":"2023-12-26T01:43:16.553237Z","shell.execute_reply":"2023-12-26T01:53:03.423704Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Download outputs","metadata":{}},{"cell_type":"code","source":"import os\nimport zipfile\n\nsound_files = [os.path.join(default_path, filename) for filename in os.listdir(default_path)]\n\n# Name of the output compressed file\nzip_filename = \"sounds.zip\"\n\n# Create a new ZIP file\nwith zipfile.ZipFile(zip_filename, \"w\") as zipf:\n    for sound_file in sound_files:\n        zipf.write(sound_file)\n        \n# Name of the upscale compressed file\n        \nfrom IPython.display import FileLink\nFileLink(zip_filename)","metadata":{"execution":{"iopub.status.busy":"2023-12-26T01:53:03.426072Z","iopub.execute_input":"2023-12-26T01:53:03.426402Z","iopub.status.idle":"2023-12-26T01:53:03.452602Z","shell.execute_reply.started":"2023-12-26T01:53:03.426357Z","shell.execute_reply":"2023-12-26T01:53:03.451455Z"},"trusted":true},"execution_count":null,"outputs":[]}]}