{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "3f6e39cd", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Populating the interactive namespace from numpy and matplotlib\n" ] } ], "source": [ "%pylab inline" ] }, { "cell_type": "code", "execution_count": 2, "id": "51ba12a5", "metadata": {}, "outputs": [], "source": [ "import os\n", "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\"" ] }, { "cell_type": "code", "execution_count": 3, "id": "629bf699", "metadata": {}, "outputs": [], "source": [ "import time\n", "import os\n", "import tqdm\n", "import shutil\n", "import random\n", "import funcy\n", "import json\n", "import numpy as np\n", "import multiprocessing\n", "import pandas as pd\n", "from suno_utils.audio import Audio\n", "from suno_utils.audio.conversion import play_audio\n", "from suno_utils.utils.podcasts import load_podcast_db, find_in_raw_feeds, load_rss_feed\n", "\n", "PODCAST_DATA_DIR = \"/mnt/data-ssd-1/data/podcasts/\"" ] }, { "cell_type": "code", "execution_count": 8, "id": "e7dca805", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "8094 found.\n" ] } ], "source": [ "podcast_df = load_podcast_db(\n", " os.path.join(PODCAST_DATA_DIR, \"meta/podcastindex_feeds.db\"), \n", " english_only=True, \n", " anchor_only=True,\n", ")\n", "podcast_df = podcast_df[podcast_df[\"language\"] == \"en-gb\"].reset_index(drop=True)\n", "found_uids = podcast_df[\"uid\"].tolist()\n", "print(len(found_uids), \"found.\")" ] }, { "cell_type": "code", "execution_count": 18, "id": "23676a33", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'14202390-6a16-51d7-b220-f8e67fa00067;b11cf160-62ca-5f3f-88e1-d9c1700f2e4f;e2f4373d-9213-5cef-9bc8-d2cfc2087dd2;d6e4a7fb-ff05-5881-81c3-7b7bd56a9368;a9c9f580-09b4-58f9-964e-e0d7c83e4b0f;2aefa7cc-c340-56e6-9575-a39df486aeee;78496542-a045-5029-907d-ed658c694628;aabcf1f4-0056-5217-b9a3-18285cacc41e;81a98de5-e0bb-590d-afc2-58aa1d503e04;c90aca91-36d1-508e-b388-b2b2b0253214;13910236-212f-58dc-bb0b-a403bbc82792;2f94f592-60bb-5d22-8f39-2e41af5f42a2;93bf77ee-c472-5e48-aae8-5ced1af6e5df;2246aaa5-4fc5-5e97-b44c-96ce12990306;69629b09-7c92-5895-9b00-446478247cd9;be317d9d-a4ca-5e66-a80b-4723d8243567;065c0dbc-58b6-58a7-8e6c-f326a2804b91;b0d5924f-e003-533a-b69d-9db8376fb36d;ec9899e1-9b5a-513d-9bcd-c7f1836b56a0;da6e5fd1-f9cf-56c0-9306-1544612c4a90;fb58caf4-ba81-5f74-99d4-bb4cefacf899;35dd72d5-0d60-5ff0-905b-544c520a3a01;8495409e-96b3-5131-9752-86a692134054;181976ea-e064-5063-80d0-cb33af15b128;a82f6892-20c9-5d4f-b3d1-400194e2002a;b6e964a4-151b-5d89-9f7b-a354b1602b00;e9fcf203-c333-5de7-a5ba-def84b1069e7;5c6d4803-e814-5678-8074-2d917de9652d;e13b0d85-336f-5822-b60b-a2139cb713b8;42454f10-8816-5890-8662-4677678b0424;249d98eb-3c69-5749-9638-3c4082c5e90e;3c81959b-c0f3-5d63-8429-0efd3ad5e3cd;6e0a8e65-6698-5a75-bd01-ca642d23d50c;327cbd32-8277-5b0a-a2c5-f8266399c27f;4603a041-35c4-5dc9-9fdb-197350f49ea4;bbd1259d-b316-54cb-acb5-c800d43ba7a7;ab462272-0ce5-57fb-8763-7af4f6d0e105;42a9d159-fdb6-5628-8242-a83503a750df;226c2135-0737-5602-9489-9e81d3de9134;e5e85257-b844-5e1f-9bb3-b81593350573;bb22f5dd-c646-5c30-9a75-71caa80c72c9;461bacc1-5fd4-579c-996c-90dccb4fb304;72900caa-13da-5c58-89f3-9b06426ed5cf;bf215731-0264-5cd8-afcc-5f442ed5e4a5;928f037b-bca6-58f6-b12c-798d0e456b2a;1705d5a0-59d2-50e1-8990-81654bfec86b;ac9760ce-f4ae-5e57-b58b-a12136a20dbc;dd8db74e-28f8-5944-95c7-ce43f704d9c2;8e8b0e91-c0d7-5d5c-baec-f605dde43b33;a606b4c8-7404-5eea-bbe2-ef4e2caa1f83'" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "random.seed(6006)\n", "random.shuffle(found_uids)\n", "\";\".join(found_uids[:50])" ] }, { "cell_type": "code", "execution_count": null, "id": "76aea14e", "metadata": {}, "outputs": [], "source": [ "python /home/georg/code/glockenspiel/suno_utils/suno_utils/scripts/fetch_podcasts.py \\\n", " --podcast_db_filepath=\"/mnt/data-ssd-1/data/podcasts/meta/podcastindex_feeds.db\" \\\n", " --podcast_ids=\"14202390-6a16-51d7-b220-f8e67fa00067;b11cf160-62ca-5f3f-88e1-d9c1700f2e4f;e2f4373d-9213-5cef-9bc8-d2cfc2087dd2;d6e4a7fb-ff05-5881-81c3-7b7bd56a9368;a9c9f580-09b4-58f9-964e-e0d7c83e4b0f;2aefa7cc-c340-56e6-9575-a39df486aeee;78496542-a045-5029-907d-ed658c694628;aabcf1f4-0056-5217-b9a3-18285cacc41e;81a98de5-e0bb-590d-afc2-58aa1d503e04;c90aca91-36d1-508e-b388-b2b2b0253214;13910236-212f-58dc-bb0b-a403bbc82792;2f94f592-60bb-5d22-8f39-2e41af5f42a2;93bf77ee-c472-5e48-aae8-5ced1af6e5df;2246aaa5-4fc5-5e97-b44c-96ce12990306;69629b09-7c92-5895-9b00-446478247cd9;be317d9d-a4ca-5e66-a80b-4723d8243567;065c0dbc-58b6-58a7-8e6c-f326a2804b91;b0d5924f-e003-533a-b69d-9db8376fb36d;ec9899e1-9b5a-513d-9bcd-c7f1836b56a0;da6e5fd1-f9cf-56c0-9306-1544612c4a90;fb58caf4-ba81-5f74-99d4-bb4cefacf899;35dd72d5-0d60-5ff0-905b-544c520a3a01;8495409e-96b3-5131-9752-86a692134054;181976ea-e064-5063-80d0-cb33af15b128;a82f6892-20c9-5d4f-b3d1-400194e2002a;b6e964a4-151b-5d89-9f7b-a354b1602b00;e9fcf203-c333-5de7-a5ba-def84b1069e7;5c6d4803-e814-5678-8074-2d917de9652d;e13b0d85-336f-5822-b60b-a2139cb713b8;42454f10-8816-5890-8662-4677678b0424;249d98eb-3c69-5749-9638-3c4082c5e90e;3c81959b-c0f3-5d63-8429-0efd3ad5e3cd;6e0a8e65-6698-5a75-bd01-ca642d23d50c;327cbd32-8277-5b0a-a2c5-f8266399c27f;4603a041-35c4-5dc9-9fdb-197350f49ea4;bbd1259d-b316-54cb-acb5-c800d43ba7a7;ab462272-0ce5-57fb-8763-7af4f6d0e105;42a9d159-fdb6-5628-8242-a83503a750df;226c2135-0737-5602-9489-9e81d3de9134;e5e85257-b844-5e1f-9bb3-b81593350573;bb22f5dd-c646-5c30-9a75-71caa80c72c9;461bacc1-5fd4-579c-996c-90dccb4fb304;72900caa-13da-5c58-89f3-9b06426ed5cf;bf215731-0264-5cd8-afcc-5f442ed5e4a5;928f037b-bca6-58f6-b12c-798d0e456b2a;1705d5a0-59d2-50e1-8990-81654bfec86b;ac9760ce-f4ae-5e57-b58b-a12136a20dbc;dd8db74e-28f8-5944-95c7-ce43f704d9c2;8e8b0e91-c0d7-5d5c-baec-f605dde43b33;a606b4c8-7404-5eea-bbe2-ef4e2caa1f83\" \\\n", " --output_dir=\"/mnt/data-ssd-1/data/podcasts/audio\" \\\n", " --last_n_episodes=1" ] }, { "cell_type": "markdown", "id": "8ea2c1cb", "metadata": {}, "source": [ "### make slices" ] }, { "cell_type": "code", "execution_count": 24, "id": "25067dc8", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "41" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from suno_utils.web.harvest import get_file_name\n", "\n", "min_episode_duration_s = 3 * 60\n", "max_episode_duration_s = 10 * 60\n", "\n", "episode_metas = []\n", "for podcast_id in found_uids[:50]:\n", " podcast_dir = os.path.join(PODCAST_DATA_DIR, \"audio\", podcast_id)\n", " for fn in os.listdir(podcast_dir):\n", " try:\n", " filepath = os.path.join(podcast_dir, fn)\n", " duration_s = Audio.get_details(filepath)[\"duration_s\"]\n", " if duration_s >= min_episode_duration_s:\n", " episode_id = get_file_name(fn)\n", " episode_metas.append({\n", " \"podcast_id\": podcast_id, \n", " \"episode_id\": episode_id, \n", " \"filepath\": filepath, \n", " \"duration_s\": np.min([duration_s, max_episode_duration_s]),\n", " })\n", " except:\n", " pass\n", "len(episode_metas)" ] }, { "cell_type": "code", "execution_count": 29, "id": "e7df0034", "metadata": {}, "outputs": [], "source": [ "SEGMENT_DIR = \"/home/georg/notebooks/datasets/great_britain/tmp/segments/\"\n", "\n", "shutil.rmtree(SEGMENT_DIR, ignore_errors=True)\n", "os.makedirs(SEGMENT_DIR, exist_ok=True)" ] }, { "cell_type": "code", "execution_count": 30, "id": "8d4ecc13", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ " 17%|█████████████████████████████▎ | 7/41 [01:30<08:50, 15.60s/it]few vad gaps were found, try increasing aggressiveness\n", " 71%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉ | 29/41 [05:36<02:22, 11.84s/it]sample rate 32000 not supported, will use infer best.\n", "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 41/41 [07:32<00:00, 11.03s/it]\n" ] } ], "source": [ "import random\n", "import tqdm\n", "from suno_utils.utils.slicer import get_segments\n", "from suno_utils.utils.numbers import safe_round\n", "import uuid\n", "\n", "segment_metas = []\n", "for episode_meta in tqdm.tqdm(episode_metas):\n", " audio = Audio.from_file(episode_meta[\"filepath\"]).get_slice(to_s=episode_meta[\"duration_s\"])\n", " episode_segments = get_segments(\n", " audio,\n", " min_gap_s=0.05,\n", " vad_level=3,\n", " min_dur_s=10,\n", " max_dur_s=30,\n", " )\n", " # discard first and last slice\n", " for segment_audio, (start_s, end_s) in episode_segments[1:-1]:\n", " segment_id = str(uuid.uuid4())\n", " segment_filepath = os.path.join(SEGMENT_DIR, segment_id) + \".mp3\"\n", " segment_audio.to_mp3(segment_filepath)\n", " segment_metas.append({\n", " \"podcast_id\": episode_meta[\"podcast_id\"], \n", " \"episode_id\": episode_meta[\"episode_id\"], \n", " \"uuid\": segment_id,\n", " \"source_audio_offset_s\": start_s,\n", " \"filepath\": segment_filepath, \n", " \"duration_s\": safe_round(end_s - start_s),\n", " })\n", "print(len(segment_metas), \"slices.\")" ] }, { "cell_type": "code", "execution_count": 32, "id": "d2c1d5e2", "metadata": {}, "outputs": [], "source": [ "# random shuffle\n", "# TODO: make stratified\n", "segment_df = pd.DataFrame(segment_metas)\n", "\n", "n_max_segments_per_episode = 5\n", "# n_max_segments_per_episode = int(segment_df.groupby([\"podcast_id\", \"episode_id\"]).count()[\"uuid\"].median())\n", "\n", "d = segment_df.groupby([\"podcast_id\", \"episode_id\"])[\"uuid\"].apply(list).to_dict()\n", "random.seed(6006)\n", "keys = []\n", "for _, v in d.items():\n", " nv = [e for e in v]\n", " random.shuffle(nv)\n", " for nnv in nv[:n_max_segments_per_episode]:\n", " keys.append(nnv)\n", "random.shuffle(keys)\n", "segment_df = segment_df.set_index(\"uuid\").loc[keys].reset_index()" ] }, { "cell_type": "code", "execution_count": null, "id": "1c8cadd1", "metadata": {}, "outputs": [], "source": [ "for _, row in segment_df.iloc[:40].iterrows():\n", " Audio.from_file(row[\"filepath\"]).play()\n", " print(row[\"uuid\"])\n", " print(\"-\"*20)" ] }, { "cell_type": "code", "execution_count": 40, "id": "0c022430", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "10" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "best_segment_ids = [\n", " \"936cf9f2-cbb6-49f8-af68-0612c815d930\",\n", " \"e9a913da-ae22-4b01-8827-68dc2d4742c1\",\n", " \"6ea8ac4b-e189-4b58-b920-4e11c8bcb3d4\",\n", " \"de693cbd-0537-4075-9284-6e2eff7b8439\",\n", " \"f407bfad-d00f-4db2-8706-ad722dbe18ba\",\n", " \"276fa7fb-8f64-4aa6-ae36-e988b69b8b00\",\n", " \"627c0a6c-203b-4e2f-9397-19f26f2d8c5a\",\n", " \"0b663cd8-d5ba-4b54-be6c-320feb6f0ce0\",\n", " \"16b2f2d6-f541-4ea4-8927-1ec8bfc6e2b5\",\n", " \"27b07b17-6258-4198-8335-8829e0bd6658\",\n", " \"19c0f64c-14fb-482d-8ceb-0a63464ef282\",\n", " \"5a518d8d-f73b-4e86-b495-020983519f5e\",\n", "]\n", "seen = set()\n", "best_clean_segment_ids = []\n", "for uid in best_segment_ids:\n", " row = segment_df[segment_df[\"uuid\"] == uid].iloc[0]\n", " if row[\"podcast_id\"] in seen:\n", " continue\n", " best_clean_segment_ids.append(uid)\n", " seen.add(row[\"podcast_id\"])\n", "len(best_clean_segment_ids)" ] }, { "cell_type": "code", "execution_count": 152, "id": "bc12d29e", "metadata": {}, "outputs": [], "source": [ "best_clean_segment_ids = [\n", " '936cf9f2-cbb6-49f8-af68-0612c815d930',\n", " 'e9a913da-ae22-4b01-8827-68dc2d4742c1',\n", " '6ea8ac4b-e189-4b58-b920-4e11c8bcb3d4',\n", " 'de693cbd-0537-4075-9284-6e2eff7b8439',\n", " 'f407bfad-d00f-4db2-8706-ad722dbe18ba',\n", " '0b663cd8-d5ba-4b54-be6c-320feb6f0ce0',\n", " '16b2f2d6-f541-4ea4-8927-1ec8bfc6e2b5',\n", " '27b07b17-6258-4198-8335-8829e0bd6658',\n", " '19c0f64c-14fb-482d-8ceb-0a63464ef282',\n", " '5a518d8d-f73b-4e86-b495-020983519f5e',\n", "]" ] }, { "cell_type": "code", "execution_count": 47, "id": "ac58dae6", "metadata": {}, "outputs": [], "source": [ "audios = []\n", "for uid in best_clean_segment_ids:\n", " row = segment_df[segment_df[\"uuid\"] == uid].iloc[0]\n", " audios.append(Audio.from_file(row[\"filepath\"]).convert(44_100, 2, 1))\n", "audio = Audio.concatenate(audios)" ] }, { "cell_type": "code", "execution_count": 48, "id": "3902f6e5", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "output_file: gb_10_comb.mp3 already exists and will be overwritten on build\n" ] } ], "source": [ "audio.to_mp3(\"gb_10_comb.mp3\")" ] }, { "cell_type": "code", "execution_count": 49, "id": "49c3b7e9", "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "audio.play()" ] }, { "cell_type": "code", "execution_count": null, "id": "50f0254f", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "edc8acf2", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "aedae77f", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "5ad29605", "metadata": {}, "source": [ "### More targeted" ] }, { "cell_type": "code", "execution_count": 98, "id": "72e54d8f", "metadata": {}, "outputs": [], "source": [ "podcast_info_df = pd.read_csv(\"/mnt/data-ssd-1/data/podcasts/meta/summaries.csv\")\n", "podcast_info_df = podcast_info_df[podcast_info_df[\"uid\"].isin(found_uids)]\n", "random.seed(6006)" ] }, { "cell_type": "code", "execution_count": 112, "id": "6a9f483c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "21 scottish found.\n" ] } ], "source": [ "scottish_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Ss]cottish\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(scottish_uids)\n", "print(len(scottish_uids), \"scottish found.\")" ] }, { "cell_type": "code", "execution_count": 113, "id": "ecd8ee57", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "6 welsh found.\n" ] } ], "source": [ "welsh_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Ww]elsh\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(welsh_uids)\n", "print(len(welsh_uids), \"welsh found.\")" ] }, { "cell_type": "code", "execution_count": 114, "id": "3eba9a50", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "12 irish found.\n" ] } ], "source": [ "irish_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Ii]rish\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(irish_uids)\n", "print(len(irish_uids), \"irish found.\")" ] }, { "cell_type": "code", "execution_count": 115, "id": "6611be4b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "5 found.\n" ] } ], "source": [ "newcastle_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Nn]ewcastle\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(newcastle_uids)\n", "print(len(newcastle_uids), \"found.\")" ] }, { "cell_type": "code", "execution_count": 116, "id": "35b35d94", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "10 found.\n" ] } ], "source": [ "liverpool_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Ll]iverpool\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(liverpool_uids)\n", "print(len(liverpool_uids), \"found.\")" ] }, { "cell_type": "code", "execution_count": 117, "id": "db2595f3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "18 found.\n" ] } ], "source": [ "yorkshire_uids = podcast_info_df[\n", " podcast_info_df[\"summary\"].str.contains(r\"\\b[Yy]orkshire\\b\", regex=True).fillna(False)\n", "][\"uid\"].tolist()\n", "random.shuffle(yorkshire_uids)\n", "print(len(yorkshire_uids), \"found.\")" ] }, { "cell_type": "code", "execution_count": 118, "id": "e4109702", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'3f5437ab-76bf-50f6-abef-8d48e77fea42;f211678d-02a9-50a4-8528-6385f433f14b;631628cb-45a6-58af-82d4-89bb07a47cd6;5e1e13de-0375-53e3-bcac-4b2c8cb99278;c2550fda-d52f-59d7-811f-3c4d5590c259;efc83b74-be2e-52a3-a034-2bf35c7faba3;e077629e-e885-5d55-a5bc-02bf95e2cd39;c9a7e827-6f1c-5caa-acca-1c60c6582d00;cacf1002-ad4b-5b5a-87d3-a79754eabd39;10ff8383-ccd5-5a63-b041-b039c67b0b2e;1f3c7754-a299-5613-bdcf-7424e333910c;ceec7168-40f8-59b1-8207-050da39ec7fe;08d885e6-ce74-5325-80e2-365ad3fa8ab6;c8903f04-4009-5acd-97f7-dbe94de85460;822b8dd4-e1dd-5be2-9047-058106507e59;8dc264d8-b676-57f4-9a72-302e596eaecb;d342c56c-7b16-59dc-a48a-7bffd82df0cb;f1814c0c-5243-598d-9a27-a0d98286c163'" ] }, "execution_count": 118, "metadata": {}, "output_type": "execute_result" } ], "source": [ "location_uids = (\n", " scottish_uids[:3] + \n", " welsh_uids[:3] + \n", " irish_uids[:3] + \n", " newcastle_uids[:3] + \n", " liverpool_uids[:3] + \n", " yorkshire_uids[:3]\n", ")\n", "\n", "location_map = {}\n", "for uid in scottish_uids:\n", " location_map[uid] = \"scottish\"\n", "for uid in welsh_uids:\n", " location_map[uid] = \"welsh\"\n", "for uid in irish_uids:\n", " location_map[uid] = \"irish\"\n", "for uid in newcastle_uids:\n", " location_map[uid] = \"newcastle\"\n", "for uid in liverpool_uids:\n", " location_map[uid] = \"liverpool\"\n", "for uid in yorkshire_uids:\n", " location_map[uid] = \"yorkshire\"\n", " \n", "\";\".join(location_uids)" ] }, { "cell_type": "code", "execution_count": null, "id": "5c353aea", "metadata": {}, "outputs": [], "source": [ "python /home/georg/code/glockenspiel/suno_utils/suno_utils/scripts/fetch_podcasts.py \\\n", " --podcast_db_filepath=\"/mnt/data-ssd-1/data/podcasts/meta/podcastindex_feeds.db\" \\\n", " --podcast_ids=\"3f5437ab-76bf-50f6-abef-8d48e77fea42;f211678d-02a9-50a4-8528-6385f433f14b;631628cb-45a6-58af-82d4-89bb07a47cd6;5e1e13de-0375-53e3-bcac-4b2c8cb99278;c2550fda-d52f-59d7-811f-3c4d5590c259;efc83b74-be2e-52a3-a034-2bf35c7faba3;e077629e-e885-5d55-a5bc-02bf95e2cd39;c9a7e827-6f1c-5caa-acca-1c60c6582d00;cacf1002-ad4b-5b5a-87d3-a79754eabd39;10ff8383-ccd5-5a63-b041-b039c67b0b2e;1f3c7754-a299-5613-bdcf-7424e333910c;ceec7168-40f8-59b1-8207-050da39ec7fe;08d885e6-ce74-5325-80e2-365ad3fa8ab6;c8903f04-4009-5acd-97f7-dbe94de85460;822b8dd4-e1dd-5be2-9047-058106507e59;8dc264d8-b676-57f4-9a72-302e596eaecb;d342c56c-7b16-59dc-a48a-7bffd82df0cb;f1814c0c-5243-598d-9a27-a0d98286c163\" \\\n", " --output_dir=\"/mnt/data-ssd-1/data/podcasts/audio\" \\\n", " --last_n_episodes=1" ] }, { "cell_type": "code", "execution_count": 121, "id": "49f3cab1", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "17" ] }, "execution_count": 121, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from suno_utils.web.harvest import get_file_name\n", "\n", "min_episode_duration_s = 3 * 60\n", "max_episode_duration_s = 10 * 60\n", "\n", "episode_metas = []\n", "for podcast_id in location_uids:\n", " podcast_dir = os.path.join(PODCAST_DATA_DIR, \"audio\", podcast_id)\n", " for fn in os.listdir(podcast_dir):\n", " try:\n", " filepath = os.path.join(podcast_dir, fn)\n", " duration_s = Audio.get_details(filepath)[\"duration_s\"]\n", " if duration_s >= min_episode_duration_s:\n", " episode_id = get_file_name(fn)\n", " episode_metas.append({\n", " \"podcast_id\": podcast_id, \n", " \"episode_id\": episode_id, \n", " \"filepath\": filepath, \n", " \"duration_s\": np.min([duration_s, max_episode_duration_s]),\n", " })\n", " except:\n", " pass\n", "len(episode_metas)" ] }, { "cell_type": "code", "execution_count": 122, "id": "90806e0c", "metadata": {}, "outputs": [], "source": [ "SEGMENT_DIR = \"/home/georg/notebooks/datasets/great_britain/tmp/segments_2/\"\n", "\n", "shutil.rmtree(SEGMENT_DIR, ignore_errors=True)\n", "os.makedirs(SEGMENT_DIR, exist_ok=True)" ] }, { "cell_type": "code", "execution_count": 123, "id": "bbfe7ada", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ " 47%|████████████████████████████████████████████████████████████████████████████████▉ | 8/17 [01:36<01:43, 11.50s/it]sample rate 32000 not supported, will use infer best.\n", " 59%|████████████████████████████████████████████████████████████████████████████████████████████████████▌ | 10/17 [02:20<02:02, 17.43s/it]sample rate 32000 not supported, will use infer best.\n", "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 17/17 [04:37<00:00, 16.35s/it]" ] }, { "name": "stdout", "output_type": "stream", "text": [ "321 slices.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] } ], "source": [ "import random\n", "import tqdm\n", "from suno_utils.utils.slicer import get_segments\n", "from suno_utils.utils.numbers import safe_round\n", "import uuid\n", "\n", "segment_metas = []\n", "for episode_meta in tqdm.tqdm(episode_metas):\n", " audio = Audio.from_file(episode_meta[\"filepath\"]).get_slice(to_s=episode_meta[\"duration_s\"])\n", " episode_segments = get_segments(\n", " audio,\n", " min_gap_s=0.05,\n", " vad_level=3,\n", " min_dur_s=10,\n", " max_dur_s=30,\n", " )\n", " # discard first and last slice\n", " for segment_audio, (start_s, end_s) in episode_segments[1:-1]:\n", " segment_id = str(uuid.uuid4())\n", " segment_filepath = os.path.join(SEGMENT_DIR, segment_id) + \".mp3\"\n", " segment_audio.to_mp3(segment_filepath)\n", " segment_metas.append({\n", " \"podcast_id\": episode_meta[\"podcast_id\"], \n", " \"episode_id\": episode_meta[\"episode_id\"], \n", " \"uuid\": segment_id,\n", " \"source_audio_offset_s\": start_s,\n", " \"filepath\": segment_filepath, \n", " \"duration_s\": safe_round(end_s - start_s),\n", " })\n", "print(len(segment_metas), \"slices.\")" ] }, { "cell_type": "code", "execution_count": 124, "id": "5755462c", "metadata": {}, "outputs": [], "source": [ "# random shuffle\n", "# TODO: make stratified\n", "segment_df = pd.DataFrame(segment_metas)\n", "\n", "n_max_segments_per_episode = 5\n", "# n_max_segments_per_episode = int(segment_df.groupby([\"podcast_id\", \"episode_id\"]).count()[\"uuid\"].median())\n", "\n", "d = segment_df.groupby([\"podcast_id\", \"episode_id\"])[\"uuid\"].apply(list).to_dict()\n", "random.seed(6006)\n", "keys = []\n", "for _, v in d.items():\n", " nv = [e for e in v]\n", " random.shuffle(nv)\n", " for nnv in nv[:n_max_segments_per_episode]:\n", " keys.append(nnv)\n", "random.shuffle(keys)\n", "segment_df = segment_df.set_index(\"uuid\").loc[keys].reset_index()" ] }, { "cell_type": "code", "execution_count": 131, "id": "f50ee7d1", "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 5a7a856d-6888-4c4f-8c7e-f8c3f2c8c08e - (822b8dd4-e1dd-5be2-9047-058106507e59)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - c3f83d04-009b-4d24-9990-8dc99604a5e5 - (c8903f04-4009-5acd-97f7-dbe94de85460)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 46f832e6-14f3-4014-977e-a84f21bde790 - (08d885e6-ce74-5325-80e2-365ad3fa8ab6)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - dfa760dc-bc01-4ccb-b091-6cac3b7f0acd - (822b8dd4-e1dd-5be2-9047-058106507e59)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 33a76863-0c60-4c24-a014-b3622ecb7a61 - (c8903f04-4009-5acd-97f7-dbe94de85460)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 2c188641-927a-42d5-9c53-9b45b07d4583 - (08d885e6-ce74-5325-80e2-365ad3fa8ab6)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 22cc44ee-ea9b-44db-9969-6ba6548bcfc1 - (c8903f04-4009-5acd-97f7-dbe94de85460)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 7d89c7f7-5575-4dba-ad73-9bfcb4a8f2a5 - (822b8dd4-e1dd-5be2-9047-058106507e59)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 2c6cbb57-e655-4507-8a06-821923673729 - (08d885e6-ce74-5325-80e2-365ad3fa8ab6)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 2691caf1-8633-44af-beb2-8a9bb93283dd - (822b8dd4-e1dd-5be2-9047-058106507e59)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 480b109b-612b-41e5-8379-6e36820dfa15 - (822b8dd4-e1dd-5be2-9047-058106507e59)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - c9d3d659-8acc-4d83-b501-17b784bec3ed - (08d885e6-ce74-5325-80e2-365ad3fa8ab6)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 1cef0ea0-31f1-442d-ac03-9095cd3f48f8 - (c8903f04-4009-5acd-97f7-dbe94de85460)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - 8409a9bf-bd62-4ea3-9e42-6540c846c7d3 - (08d885e6-ce74-5325-80e2-365ad3fa8ab6)\n", "--------------------\n" ] }, { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "liverpool - ad432b6c-7b39-4570-9fcc-94489dab2255 - (c8903f04-4009-5acd-97f7-dbe94de85460)\n", "--------------------\n" ] } ], "source": [ "n = 0\n", "for _, row in segment_df.iterrows():\n", " if location_map[row[\"podcast_id\"]] != \"liverpool\":\n", " continue\n", " Audio.from_file(row[\"filepath\"]).play()\n", " print(location_map[row[\"podcast_id\"]], \"-\", row[\"uuid\"], \"- ({})\".format(row[\"podcast_id\"]))\n", " print(\"-\"*20)\n", " n += 1\n", " if n == 20:\n", " break" ] }, { "cell_type": "code", "execution_count": null, "id": "52fe0b85", "metadata": {}, "outputs": [], "source": [ "# liverpool, \"yorkshire\"" ] }, { "cell_type": "code", "execution_count": 132, "id": "ab7b22c2", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "9" ] }, "execution_count": 132, "metadata": {}, "output_type": "execute_result" } ], "source": [ "best_segment_ids = [\n", " \"7eff72f1-7ef3-41a9-a12b-6c811938cb86\", # scottish\n", " \"bf54d477-dbb6-4028-87d6-6ffc4c7e39f6\", # scottish\n", " \"10dd40e1-7920-4799-aa49-54dd512ae7b8\", # scottish\n", " \"3ecc8d93-975e-4167-bf0b-b70d69671682\", # irish\n", " \"b1269ace-5dcc-4a63-876a-889c711eebfb\", # irish\n", " \"38a44ab8-db38-4908-a4fd-50d461e751b1\", # irish\n", " \"c5783b12-90df-4534-9956-3b410182ae50\", # welsh\n", " \"8a4788ba-1ffd-42b7-bdaa-e05bc3724a14\", # welsh\n", " \"17fa2ceb-4878-4ff7-bb48-f2bd66ba46e2\", # newcastle\n", " \"d1829b06-f7bc-4d4d-844b-cb9a850c1544\", # newcastle\n", " \"574c406b-d041-4eda-bc87-8be5c3f07847\", # newcastle\n", "\n", "]\n", "seen = set()\n", "best_clean_segment_ids = []\n", "for uid in best_segment_ids:\n", " row = segment_df[segment_df[\"uuid\"] == uid].iloc[0]\n", " if row[\"podcast_id\"] in seen:\n", " continue\n", " best_clean_segment_ids.append(uid)\n", " seen.add(row[\"podcast_id\"])\n", "len(best_clean_segment_ids)" ] }, { "cell_type": "code", "execution_count": 134, "id": "015e3692", "metadata": {}, "outputs": [], "source": [ "best_clean_segment_ids = [\n", " '7eff72f1-7ef3-41a9-a12b-6c811938cb86',\n", " 'bf54d477-dbb6-4028-87d6-6ffc4c7e39f6',\n", " '10dd40e1-7920-4799-aa49-54dd512ae7b8',\n", " '3ecc8d93-975e-4167-bf0b-b70d69671682',\n", " 'b1269ace-5dcc-4a63-876a-889c711eebfb',\n", " '38a44ab8-db38-4908-a4fd-50d461e751b1',\n", " 'c5783b12-90df-4534-9956-3b410182ae50',\n", " '17fa2ceb-4878-4ff7-bb48-f2bd66ba46e2',\n", " 'd1829b06-f7bc-4d4d-844b-cb9a850c1544',\n", "]" ] }, { "cell_type": "code", "execution_count": 135, "id": "ea8fe8e2", "metadata": {}, "outputs": [], "source": [ "audios = []\n", "for uid in best_clean_segment_ids:\n", " row = segment_df[segment_df[\"uuid\"] == uid].iloc[0]\n", " audios.append(Audio.from_file(row[\"filepath\"]).convert(44_100, 2, 1))\n", "audio = Audio.concatenate(audios)" ] }, { "cell_type": "code", "execution_count": 136, "id": "e90d0d47", "metadata": {}, "outputs": [], "source": [ "audio.to_mp3(\"gb_10_comb_2.mp3\")" ] }, { "cell_type": "code", "execution_count": 137, "id": "508e923a", "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "audio.play()" ] }, { "cell_type": "code", "execution_count": 151, "id": "1f87eb7e", "metadata": {}, "outputs": [], "source": [ "# !zip -r blabla.zip tmp/blabla" ] }, { "cell_type": "code", "execution_count": null, "id": "e7123a5d", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "f4690de1", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "bc19c118", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "768574a7", "metadata": {}, "source": [ "#### get from rev" ] }, { "cell_type": "code", "execution_count": 69, "id": "f57638dd", "metadata": {}, "outputs": [], "source": [ "from suno_utils.web.rev import get_auth_string, parse_transcript, get_finished_order, get_transcript" ] }, { "cell_type": "code", "execution_count": 70, "id": "026d1f27", "metadata": {}, "outputs": [], "source": [ "with open(\"/home/georg/.secrets/secrets.json\") as f:\n", " secrets = json.load(f)\n", " \n", "CLIENT_API_KEY = secrets[\"REV_CLIENT_API_KEY\"]\n", "USER_API_KEY = secrets[\"REV_USER_API_KEY\"]\n", "AUTH_STR = get_auth_string(CLIENT_API_KEY, USER_API_KEY)" ] }, { "cell_type": "code", "execution_count": 138, "id": "704377a1", "metadata": {}, "outputs": [], "source": [ "# order_number = \"TC0389644593\"\n", "order_number = \"TC0100769719\"\n", "transcript_meta = get_finished_order(AUTH_STR, order_number)" ] }, { "cell_type": "code", "execution_count": 139, "id": "6c3aa885", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Exception in thread QueueFeederThread:\n", "Traceback (most recent call last):\n", " File \"/usr/lib/python3.8/multiprocessing/queues.py\", line 235, in _feed\n", " close()Exception ignored in: \n", "\n", "Traceback (most recent call last):\n", " File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 132, in __del__\n", " File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 177, in close\n", " self._close()\n", " File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 361, in _close\n", " _close(self._handle)\n", "OSError: [Errno 9] Bad file descriptor\n", "self._close()\n", " File \"/usr/lib/python3.8/multiprocessing/connection.py\", line 361, in _close\n", " _close(self._handle)\n", "OSError: [Errno 9] Bad file descriptor\n", "\n", "During handling of the above exception, another exception occurred:\n", "\n", "Traceback (most recent call last):\n", " File \"/usr/lib/python3.8/threading.py\", line 932, in _bootstrap_inner\n", " self.run()\n", " File \"/usr/lib/python3.8/threading.py\", line 870, in run\n", " self._target(*self._args, **self._kwargs)\n", " File \"/usr/lib/python3.8/multiprocessing/queues.py\", line 266, in _feed\n", " queue_sem.release()\n", "ValueError: semaphore or lock released too many times\n" ] } ], "source": [ "transcript_id = transcript_meta[0][\"rev_id\"]\n", "transcript = get_transcript(AUTH_STR, transcript_id)\n", "parsed_transcript = parse_transcript(transcript)" ] }, { "cell_type": "code", "execution_count": 140, "id": "b0f15464", "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Speaker 1\n", "And it was such a pleasure to talk with him about what's happening in his life at the moment and his journey as a musician, as well as taking some time to reflect on events of the moment, which was really fascinating and really worthwhile. So, thank you so much to Matthew for joining me for that. That was great. I loved it. I really did. I also had a really good conversation with a gentleman called Stuart Beckham, who runs a website called --\n", "\n", "Speaker 2\n", "I support colleagues with any issues they may be facing. Our work is far from over but so far, we've delivered some meaningful improvements with many more plans. Today, I'm joined by Keeper of the Registers of Scotland's Jennifer Henderson, Non-Executive Director at Vos, Andrew Harvey, and our EDI Lead at Vos, Elizabeth Akurang. Thank you so much for joining us today.\n", "\n", "Speaker 3\n", "The weekly shout-out. If you have any suggestions for people or projects that you think we should promote, let us know via our socials or send me an email. That being said, this week's shout-out goes to the newest member of Team SGN, Duncan Robertson. He's done some really cool work before joining us, working on his own YouTube channel, Game Coping, as well as other projects. I'd like to take this opportunity to publicly welcome Duncan to the team and ask if you guys watching could do the same.\n", "\n", "Speaker 4\n", "Show us exactly where we are as a group, I think. So it'll be a good test, obviously, away from home and, um, one we're all looking forward to though, but, yeah, hopefully we just keep improving week on week and hopefully that gets us, uh, to where we want to get at the end of the season.\n", "\n", "Speaker 5\n", "Brilliant. Okay, I, I'll come back to the rugby a little bit later.\n", "\n", "Speaker 4\n", "Mm-hmm.\n", "\n", "Speaker 5\n", "But now, I've had a bit of a dive in your Instagram.\n", "\n", "Speaker 4\n", "Mm-hmm.\n", "\n", "Speaker 5\n", "Been doing some scrolling. Got a few questions for you, to be honest with you. So we'll start off with the nice stuff.\n", "\n", "Speaker 4\n", "Mm-hmm.\n", "\n", "Speaker 5\n", "So Zeus and Willow are your dogs.\n", "\n", "Speaker 4\n", "Yeah.\n", "\n", "Speaker 5\n", "They feature heavily on the Gram.\n", "\n", "Speaker 6\n", "Uh, [inaudible] subject we're going to talk about today, very close to my heart [inaudible] little bit hard that you probably [inaudible] my voice. It's been quite busy. [inaudible] and the [inaudible] we still get out and about [inaudible] enjoying the lovely weather that we're getting in the UK, at the moment.\n", "\n", "Speaker 7\n", "I ask God to give me [inaudible] after his own heart. See this hard thing is very important for me, especially now with the war in Ukraine and Jews [inaudible] we've got to be attracted towards that. And what I find fascinating is that there's a hell of a lot of [inaudible] around about me [inaudible] the COVID and this war thing.\n", "\n", "Speaker 8\n", "Um --\n", "\n", "Speaker 9\n", "We'll come to that in a bit. We'll come to that in a bit.\n", "\n", "Speaker 8\n", "And, um, yeah. That's where we were born. There was a group of us. I think, I'm not sure, I, if I'd even done my first triathlon at that point, maybe one. But there was a group that have never done a triathlon. There were some that were [inaudible] so that was great [inaudible]\n", "\n", "Speaker 9\n", "I was going to ask. So it's, it happened 15 years ago, now. [inaudible] And just over 15 years ago and that was when you first met in, in surgery, but, but prior to that --\n", "\n", "Speaker 10\n", "Um, [inaudible] I thank you for your Australian $10. Uh, good morning. Can't wait for the graphics, Alex. So even people in Australia are getting up at early hours of the morning on a Tuesday to, you know, to watch your graphics, Alex. It's, it's incredible. Um, [inaudible] Nick. Uh, this is Nick. Evening, guys. [inaudible] being stuck at work for 10 hours. [inaudible] and lost a tenner. What a day. Um, oh dear.\n", "\n", "Speaker 9\n", "Can only get better.\n", "\n", "Speaker 10\n", "It can. Uh --\n", "\n", "Speaker 11\n", "The club doing great things, the change of ownership, the change of management, et cetera, et cetera. Um, but, like I said, things are requiring patience at the moment and it's, it's difficult. It's like do you make a podcast as well a -- and then something breaks the next day and it's sort of like undoes that sort of stuff and we're, we're like everybody else, we're expecting an announcement to deal with, with the manager, um, and it's just, um, it's just not cool.\n", "\n", "Speaker 12\n", "No. It's not [inaudible] like --\n", "\n" ] } ], "source": [ "for e in parsed_transcript.speaker_turns:\n", " print(e[\"speaker_id\"])\n", " print(e[\"text\"])\n", " print()" ] }, { "cell_type": "code", "execution_count": null, "id": "f5339351", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "86ff3636", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "3c23c8cb", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "58fc4c3e", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "1571b2c9", "metadata": {}, "source": [ "## Playground" ] }, { "cell_type": "code", "execution_count": 13, "id": "4edf63ee", "metadata": {}, "outputs": [], "source": [ "for uid in found_uids:\n", " with open(os.path.join(PODCAST_DATA_DIR, f\"bulk_rss_feeds/parsed_feeds/{uid}.json\")) as f:\n", " rss_info = json.load(f)\n", " break" ] }, { "cell_type": "code", "execution_count": null, "id": "9e34d902", "metadata": {}, "outputs": [], "source": [ "# TODO: all rss_info[\"uid\"] is the same" ] }, { "cell_type": "code", "execution_count": 14, "id": "4c5aa037", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'uid': '56742629-8164-5534-b23a-a970818e376d',\n", " 'language': 'en-gb',\n", " 'title': 'Tea And A Butty',\n", " 'subtitle': 'We\\'re Dominic and Erica, a British and American husband and wife podcast that chats about topics you might discuss with a friend over Tea and a Butty. Support this podcast: https://anchor.fm/teaandabutty/supporthttps://anchor.fm/teaandabutty/support\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
uidtitlesummarytagsauthorauthor_email
0e71645e6-b516-5d3e-b3bc-dd3b9ebd6d6cWest Coast BallersNonesportsWest Coast Ballerswestcoastballers27@gmail.com
17b9047de-4daf-55dc-874e-b9ff7e729575Gab en vivoGab colabora en diversos medios.improv;comedy;improvLa Producción Simbólicafeeds@spreaker.com
\n", "" ], "text/plain": [ " uid title \\\n", "0 e71645e6-b516-5d3e-b3bc-dd3b9ebd6d6c West Coast Ballers \n", "1 7b9047de-4daf-55dc-874e-b9ff7e729575 Gab en vivo \n", "\n", " summary tags \\\n", "0 None sports \n", "1 Gab colabora en diversos medios. improv;comedy;improv \n", "\n", " author author_email \n", "0 West Coast Ballers westcoastballers27@gmail.com \n", "1 La Producción Simbólica feeds@spreaker.com " ] }, "execution_count": 90, "metadata": {}, "output_type": "execute_result" } ], "source": [ "summary_df.head(2)" ] }, { "cell_type": "code", "execution_count": null, "id": "40848248", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "adeb6ec0", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "97b0b993", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "7d12125f", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "20bd3ff8", "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": "3.8.10" } }, "nbformat": 4, "nbformat_minor": 5 }