{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "80c06cf2",
"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": "bbc054a8",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\""
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "ec8e0bb6",
"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 (\n",
" load_podcast_db, find_in_raw_feeds, load_rss_feed, multicore_apply, _load_rss_text, get_file_name,\n",
" _clean_episode_url\n",
")"
]
},
{
"cell_type": "markdown",
"id": "a06dac88",
"metadata": {},
"source": [
"### Check CC0 podcasts"
]
},
{
"cell_type": "code",
"execution_count": 56,
"id": "c357b9c2",
"metadata": {},
"outputs": [],
"source": [
"PODCAST_DATA_DIR = \"/mnt/data-ssd-1/data/podcasts/\"\n",
"\n",
"podcast_df = load_podcast_db(\n",
" os.path.join(PODCAST_DATA_DIR, \"meta/podcastindex_feeds.db\"), english_only=True, anchor_only=False\n",
")\n",
"summaries_df = pd.read_csv(os.path.join(PODCAST_DATA_DIR, \"meta/summaries.csv\"))"
]
},
{
"cell_type": "code",
"execution_count": 62,
"id": "fe69f334",
"metadata": {},
"outputs": [],
"source": [
"# filter out bad uuids\n",
"# group by podcast and keep only with min 5 episodes\n",
"episodes_df = pd.read_csv(\"/mnt/data-ssd-1/data/podcasts/cc_data/cc_episodes.csv\")\n",
"episodes_df = episodes_df[episodes_df[\"uuid\"].str.len() == 36]\n",
"episodes_df[\"podcast_id\"] = episodes_df[\"podcast_id\"].map(podcast_df.set_index(\"id\")[\"uid\"].to_dict())\n",
"episodes_df = episodes_df.dropna(subset=[\"podcast_id\"])\n",
"proposed_podcast_ids = set(episodes_df[\"podcast_id\"].value_counts().loc[lambda x: x >= 5].index)\n",
"episodes_df = episodes_df[episodes_df[\"podcast_id\"].isin(proposed_podcast_ids)]"
]
},
{
"cell_type": "code",
"execution_count": 67,
"id": "6927569e",
"metadata": {},
"outputs": [],
"source": [
"# assemble info and exclude things like 'music' and anchor emails\n",
"info_df = summaries_df[\n",
" summaries_df[\"uid\"].isin(set(episodes_df[\"podcast_id\"])) &\n",
" ~summaries_df[\"tags\"].fillna(\"\").str.contains(\"music\")\n",
"].reset_index(drop=True)\n",
"# add info like n episodes, last episode URL etc\n",
"tmp_map = podcast_df.drop_duplicates(subset=[\"uid\"], keep=False).set_index(\"uid\")\n",
"info_df = info_df[info_df[\"uid\"].isin(set(tmp_map.index))].reset_index(drop=True)\n",
"info_df[\"rss_url\"] = info_df[\"uid\"].map(tmp_map[\"url\"])\n",
"info_df[\"link\"] = info_df[\"uid\"].map(tmp_map[\"link\"])\n",
"info_df[\"n_episodes\"] = info_df[\"uid\"].map(tmp_map[\"episode_count\"])\n",
"info_df[\"newest_episode_pubdate\"] = info_df[\"uid\"].map(tmp_map[\"newest_item_pubdate\"])\n",
"info_df[\"oldest_episode_pubdate\"] = info_df[\"uid\"].map(tmp_map[\"oldest_item_pubdate\"])\n",
"info_df = info_df[info_df[\"newest_episode_pubdate\"].dt.year >= 2018]\n",
"h = episodes_df.sort_values(\n",
" by=\"published_date\", ascending=False\n",
").groupby(\"podcast_id\")[\"audio_url\"].apply(list).str[:3]\n",
"info_df[\"episode_url_0\"] = h.loc[info_df[\"uid\"].values].str[0].values\n",
"info_df[\"episode_url_1\"] = h.loc[info_df[\"uid\"].values].str[1].values\n",
"info_df[\"episode_url_2\"] = h.loc[info_df[\"uid\"].values].str[2].values\n",
"info_df = info_df.reset_index(drop=True)\n",
"print(info_df.shape[0], \"good candidates found.\")"
]
},
{
"cell_type": "code",
"execution_count": 97,
"id": "08d2db40",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"House on the Rock Family Church\n",
"Sunday and Wednesday Sermons from House on the Rock Family Church in Wind Gap, PA.\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"row = info_df.iloc[0]\n",
"print(row[\"title\"])\n",
"print(row[\"summary\"])\n",
"play_audio(row[\"episode_url_0\"])"
]
},
{
"cell_type": "code",
"execution_count": 98,
"id": "93b3de89",
"metadata": {},
"outputs": [],
"source": [
"info_df.to_csv(\"tmp/cc_podcasts.csv\", index=False)"
]
},
{
"cell_type": "markdown",
"id": "6f13f8b1",
"metadata": {},
"source": [
"#### Annotate"
]
},
{
"cell_type": "code",
"execution_count": 99,
"id": "897d4c0f",
"metadata": {},
"outputs": [],
"source": [
"annotate_df = pd.read_csv(\"tmp/cc_podcasts.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 100,
"id": "ac569f57",
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"from suno_utils.pipeline.labeler import SunoAPIClient\n",
"\n",
"with open(\"/home/georg/.secrets/secrets.json\") as f:\n",
" SUNO_API_TOKEN = json.load(f)[\"suno_api_token\"]\n",
"\n",
"suno_client = SunoAPIClient(token=SUNO_API_TOKEN) "
]
},
{
"cell_type": "code",
"execution_count": 101,
"id": "5a1df24e",
"metadata": {},
"outputs": [],
"source": [
"def get_task_metadata(uid, audio_url_0, audio_url_1, audio_url_2, title, description, url, n_episodes):\n",
" return {\n",
" \"blocks\": [\n",
" {\n",
" \"type\": \"audio\",\n",
" \"label\": \"Audio Sample 1:\",\n",
" \"data\": {\"value\": None if pd.isnull(audio_url_0) else audio_url_0},\n",
" }, \n",
" {\n",
" \"type\": \"audio\",\n",
" \"label\": \"Audio Sample 2:\",\n",
" \"data\": {\"value\": None if pd.isnull(audio_url_1) else audio_url_1},\n",
" }, \n",
" {\n",
" \"type\": \"audio\",\n",
" \"label\": \"Audio Sample 3:\",\n",
" \"data\": {\"value\": None if pd.isnull(audio_url_2) else audio_url_2},\n",
" }, \n",
" {\n",
" \"type\": \"markdown\",\n",
" \"data\": {\n",
" \"value\": (\n",
" f\"(id:{uid})\\n\\n - {title}\\n\\n{description}\\n\\n{url}\\n\\n\"\n",
" f\"Total Episodes: {n_episodes}\"\n",
" )\n",
" },\n",
" }, \n",
" {\n",
" \"type\": \"radio\",\n",
" \"label\": \"The podcast contains\",\n",
" \"data\": {\"key\": \"contentType\"},\n",
" \"props\": {\n",
" \"options\": [\n",
" {\"value\": v} \n",
" for v in [\n",
" \"1 Male - single speaker\",\n",
" \"1 Male - multi speaker\",\n",
" \"1 Female - single speaker\",\n",
" \"1 Female - multi speaker\",\n",
" \"Multiple - multi speaker\",\n",
" \"Not usable\",\n",
" ]\n",
" ],\n",
" \"direction\": \"column\",\n",
" },\n",
" },\n",
" ],\n",
" \"podcast_id\": uid,\n",
" }\n",
"\n",
"work_items = [\n",
" {\n",
" \"metadata\": get_task_metadata(\n",
" row[\"uid\"], \n",
" row[\"episode_url_0\"],\n",
" row[\"episode_url_1\"],\n",
" row[\"episode_url_2\"],\n",
" row[\"title\"], \n",
" row[\"summary\"], \n",
" row[\"link\"], \n",
" row[\"n_episodes\"],\n",
" ),\n",
" } \n",
" for _, row in annotate_df.iterrows()\n",
"]"
]
},
{
"cell_type": "code",
"execution_count": 102,
"id": "42ad5385",
"metadata": {},
"outputs": [],
"source": [
"task_name = \"CC EN Podcasts\"\n",
"work_item_set = suno_client.create_work_item_set(name=task_name)\n",
"remote_work_items = suno_client.create_work_items(work_items, work_item_set_id=work_item_set.id)\n",
"with open(\"tmp/{}.json\".format(task_name.lower().replace(\" \", \"_\")), \"w\") as f:\n",
" json.dump(remote_work_items, f)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "6cead153",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "3c9ea364",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "24aee7e5",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"id": "63723b6e",
"metadata": {},
"source": [
"### assemble output"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a6bcd80a",
"metadata": {},
"outputs": [],
"source": [
"annotate_df = pd.read_csv(\"tmp/cc_podcasts.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 104,
"id": "41164bc2",
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"from suno_utils.pipeline.labeler import SunoAPIClient\n",
"\n",
"with open(\"/home/georg/.secrets/secrets.json\") as f:\n",
" SUNO_API_TOKEN = json.load(f)[\"suno_api_token\"]\n",
"\n",
"suno_client = SunoAPIClient(token=SUNO_API_TOKEN) "
]
},
{
"cell_type": "code",
"execution_count": 106,
"id": "8f215a92",
"metadata": {},
"outputs": [],
"source": [
"from suno_utils.pipeline.labeler import WorkItemSet\n",
"\n",
"labeled_results = WorkItemSet(\n",
" id=\"23bf29a4-74f5-4ee9-b0c2-151951d6967c\", name=\"CC EN Podcasts\", client=suno_client\n",
").get_results()"
]
},
{
"cell_type": "code",
"execution_count": 109,
"id": "17d91544",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"3 relevant podcasts found\n",
"2 relevant podcasts found\n",
"1 relevant podcasts found\n",
"0 relevant podcasts found\n",
"0 relevant podcasts found\n"
]
}
],
"source": [
"s1 = set()\n",
"s2 = set()\n",
"s3 = set()\n",
"s4 = set()\n",
"s5 = set()\n",
"for work_item in labeled_results[\"workItems\"]:\n",
" podcast_id = work_item[\"metadata\"][\"podcast_id\"]\n",
" for annotation in work_item[\"assignments\"]:\n",
" if annotation[\"data\"].get(\"contentType\") == \"1 Male - single speaker\":\n",
" s1.add(podcast_id)\n",
" if annotation[\"data\"].get(\"contentType\") == \"1 Male - multi speaker\":\n",
" s2.add(podcast_id)\n",
" if annotation[\"data\"].get(\"contentType\") == \"1 Female - single speaker\":\n",
" s3.add(podcast_id)\n",
" if annotation[\"data\"].get(\"contentType\") == \"1 Female - multi speaker\":\n",
" s4.add(podcast_id)\n",
" if annotation[\"data\"].get(\"contentType\") == \"Multiple - multi speaker:\":\n",
" s5.add(podcast_id)\n",
"print(len(s1), \"relevant podcasts found\")\n",
"print(len(s2), \"relevant podcasts found\")\n",
"print(len(s3), \"relevant podcasts found\")\n",
"print(len(s4), \"relevant podcasts found\")\n",
"print(len(s5), \"relevant podcasts found\")"
]
},
{
"cell_type": "code",
"execution_count": 116,
"id": "d83df0c7",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(male) - 3fe70250-1706-5320-a038-2fb0bd600a69\n",
"Title: SCP Un[REDACTED\n",
"Link: http://kelseynarrates.blogspot.com/\n",
"Episode sample 1: https://scpnarration.s3.us-east-2.amazonaws.com/UnREDACTED_SCP-5602_-_Why_Did_The_SCP_Cross_The_Road_.mp3\n",
"Episode sample 2: https://scpnarration.s3.us-east-2.amazonaws.com/UnREDACTED_SCP-5574_-_Imaginary_Suffering.mp3\n",
"Episode sample 3: https://scpnarration.s3.us-east-2.amazonaws.com/GRANT_REQUEST_FOR_THE_CYBERNETIC_MODIFICATION_OF_THE_CORPSE_OF_THE_CROCOTEUTHIS_GIGANTIS.mp3\n",
"\n",
"(male) - af08bf5a-798d-5f44-a843-f99b6cd02650\n",
"Title: House on the Rock Family Church\n",
"Link: http://www.jellycast.com\n",
"Episode sample 1: https://hotrfc.jellycast.com/files/audio/02%201%20Peter%20Part%202.mp3\n",
"Episode sample 2: https://hotrfc.jellycast.com/files/audio/03%20Together%20With%20One%20Accord%20In%20One%20Place%20Part%203.mp3\n",
"Episode sample 3: https://hotrfc.jellycast.com/files/audio/01%20The%20Church%20-%20The%20Challenge%20From%20Without.mp3\n",
"\n",
"(male) - aa11479f-86d4-5ba6-baee-6bfa02421a96\n",
"Title: True Life Church\n",
"Link: http://www.truelife.church\n",
"Episode sample 1: https://theconnection.jellycast.com/files/audio/podcast%205.15.2022.mp3\n",
"Episode sample 2: https://theconnection.jellycast.com/files/audio/Podcast%205.8.mp3\n",
"Episode sample 3: https://theconnection.jellycast.com/files/audio/Podcast%205.1.22.mp3\n",
"\n",
"(male-multi) - 2998862c-f0d7-58e5-a18f-759683029bb9\n",
"Title: That's Life Podcast\n",
"Link: http://www.ratthaus.net/tlpodcast/\n",
"Episode sample 1: https://static1.squarespace.com/static/54232058e4b04580028beded/t/60613b22aad1991a031c38d0/1616984916217/TL+ep+220.mp3\n",
"Episode sample 2: https://static1.squarespace.com/static/54232058e4b04580028beded/t/60613a6daad1991a031c2f1f/1616984731381/TL+ep+219.mp3\n",
"Episode sample 3: https://static1.squarespace.com/static/54232058e4b04580028beded/t/606138785ded127ce82c6e95/1616984264560/TL+ep+218.mp3\n",
"\n",
"(male-multi) - beff1938-5b27-5868-a4ad-efdf26b3a781\n",
"Title: The Frustrated Fans Podcast\n",
"Link: http://www.frustratedpodcast.com\n",
"Episode sample 1: https://frustratedpodcast.jellycast.com/files/audio/Frustrated%20Fan%20Ep%2075.mp3\n",
"Episode sample 2: https://frustratedpodcast.jellycast.com/files/audio/Recorder%20Episode%2005.mp3\n",
"Episode sample 3: https://frustratedpodcast.jellycast.com/files/audio/Frustrated%20Fan%20Ep%2074.mp3\n",
"\n",
"(female) - 2de2084e-c60f-5a39-b3a4-6aefafa0201f\n",
"Title: Forgotten Classics\n",
"Link: http://hcforgottenclassics.blogspot.com/\n",
"Episode sample 1: http://www.stamedia.org/EP373_House21-23.mp3\n",
"Episode sample 2: http://www.stamedia.org/EP375_House-finale.mp3\n",
"Episode sample 3: http://www.stamedia.org/EP374_House24-26.mp3\n",
"\n"
]
}
],
"source": [
"for uid in s1:\n",
" print(\"(male) -\", uid)\n",
" df = annotate_df[annotate_df['uid'] == uid].iloc[0]\n",
" print(\"Title:\", df[\"title\"])\n",
" print(\"Link:\", df[\"link\"])\n",
" print(\"Episode sample 1:\", df[\"episode_url_0\"])\n",
" print(\"Episode sample 2:\", df[\"episode_url_1\"])\n",
" print(\"Episode sample 3:\", df[\"episode_url_2\"])\n",
" print()\n",
"for uid in s2:\n",
" print(\"(male-multi) -\", uid)\n",
" df = annotate_df[annotate_df['uid'] == uid].iloc[0]\n",
" print(\"Title:\", df[\"title\"])\n",
" print(\"Link:\", df[\"link\"])\n",
" print(\"Episode sample 1:\", df[\"episode_url_0\"])\n",
" print(\"Episode sample 2:\", df[\"episode_url_1\"])\n",
" print(\"Episode sample 3:\", df[\"episode_url_2\"])\n",
" print()\n",
"for uid in s3:\n",
" print(\"(female) -\", uid)\n",
" df = annotate_df[annotate_df['uid'] == uid].iloc[0]\n",
" print(\"Title:\", df[\"title\"])\n",
" print(\"Link:\", df[\"link\"])\n",
" print(\"Episode sample 1:\", df[\"episode_url_0\"])\n",
" print(\"Episode sample 2:\", df[\"episode_url_1\"])\n",
" print(\"Episode sample 3:\", df[\"episode_url_2\"])\n",
" print()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "002017dd",
"metadata": {},
"outputs": [],
"source": [
"# outreach_df = outreach_df[outreach_df[\"id\"].isin(usable_podcast_id)]\n",
"# outreach_df = outreach_df.reset_index(drop=True)\n",
"# print(outreach_df.shape[0], \"podcasts found\")\n",
"# outreach_df.head(2)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "4f82ee88",
"metadata": {},
"outputs": [],
"source": [
"# outreach_df.to_cssv(\"outreach_en_phl.csv\", index=False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e2553a52",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "24188be5",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "934ddef3",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "edd98953",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "640b3a7a",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"id": "7e45a5eb",
"metadata": {},
"source": [
"## Make samples"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "3fb4aacb",
"metadata": {},
"outputs": [],
"source": [
"# (female) - 2de2084e-c60f-5a39-b3a4-6aefafa0201f - Forgotten Classics\n",
"a = Audio.from_url(\"http://www.stamedia.org/EP373_House21-23.mp3\")\n",
"podcast_1_sample_1 = a.get_slice(2*60+45, 3*60+15)\n",
"podcast_1_sample_2 = a.get_slice(18*60+32, 19*60+2)\n",
"\n",
"# (female) - Brain Science\n",
"a = Audio.from_file(\"/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/191-BS-Review.mp3\")\n",
"podcast_2_sample_1 = a.get_slice(10*60, 10*60+30)\n",
"podcast_2_sample_2 = a.get_slice(50*60, 50*60+30)\n",
"\n",
"# (male) - Creative Haven\n",
"a = Audio.from_url(\"https://traffic.libsyn.com/secure/creativehaven/92_-_Bill_Small.mp3\")\n",
"podcast_3_sample_1 = a.get_slice(13*60+11, 13*60+41)\n",
"podcast_3_sample_2 = a.get_slice(29*60+30, 30*60+0)\n",
"\n",
"# (male) - The Greatest Story of All Time\n",
"a = Audio.from_url(\"https://d3ctxlq1ktw2nl.cloudfront.net/staging/2022-3-10/258985649-44100-2-3757f056abd92.m4a\")\n",
"podcast_4_sample_1 = a.get_slice(1*60, 1*60+30)\n",
"podcast_4_sample_2 = a.get_slice(10*60, 10*60+30)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "c810f2e0",
"metadata": {},
"outputs": [],
"source": [
"!mkdir sample_segments"
]
},
{
"cell_type": "code",
"execution_count": 39,
"id": "fe9674f8",
"metadata": {},
"outputs": [],
"source": [
"podcast_1_sample_1.to_mp3(\"sample_segments/voice_1_sample_1.mp3\")\n",
"podcast_1_sample_2.to_mp3(\"sample_segments/voice_1_sample_2.mp3\")\n",
"podcast_2_sample_1.to_mp3(\"sample_segments/voice_2_sample_1.mp3\")\n",
"podcast_2_sample_2.to_mp3(\"sample_segments/voice_2_sample_2.mp3\")\n",
"podcast_3_sample_1.to_mp3(\"sample_segments/voice_3_sample_1.mp3\")\n",
"podcast_3_sample_2.to_mp3(\"sample_segments/voice_3_sample_2.mp3\")\n",
"podcast_4_sample_1.to_mp3(\"sample_segments/voice_4_sample_1.mp3\")\n",
"podcast_4_sample_2.to_mp3(\"sample_segments/voice_4_sample_2.mp3\")"
]
},
{
"cell_type": "code",
"execution_count": 40,
"id": "f32920a6",
"metadata": {},
"outputs": [],
"source": [
"!tar -czf sample_segments.tar.gz sample_segments"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a735c2d9",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "47fa16d8",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "25f73dda",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"id": "1925ec38",
"metadata": {},
"source": [
"### Check brainscience"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "52303732",
"metadata": {},
"outputs": [],
"source": [
"AUDIO_DIR = \"/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/\""
]
},
{
"cell_type": "code",
"execution_count": 30,
"id": "2934d030",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/6-bsp-FirstIdea.mp3',\n",
" '/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/2-bsp-OnIntelligence.mp3']"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"fps = [os.path.join(AUDIO_DIR, fn) for fn in os.listdir(AUDIO_DIR) if \".mp3\" in fn]\n",
"random.seed(6006)\n",
"random.shuffle(fps)\n",
"fps[:2]"
]
},
{
"cell_type": "code",
"execution_count": 31,
"id": "3cfdfc10",
"metadata": {},
"outputs": [],
"source": [
"repeat_guests = [l.split()[0] for l in \"\"\"\n",
"grant 5\n",
"hawkins 5\n",
"churchland 3\n",
"graziano 3\n",
"ratey 3\n",
"koch 3\n",
"fields 3\n",
"medina 2\n",
"buzsaki 2\n",
"goldberg 2\n",
"benedetti 2\n",
"eagleman 2\n",
"macknik 2\n",
"doidge 2\n",
"chemero 2\n",
"sporns 2\n",
"milner 2\n",
"damasio 2\n",
"burton 2\n",
"marder 2\n",
"gazzaniga 2\n",
"baroncohen 2\n",
"merzenich 2\n",
"thompson 2\n",
"panksepp 2\n",
"herz 2\n",
"wolf 2\n",
"\"\"\".strip().split(\"\\n\")]"
]
},
{
"cell_type": "code",
"execution_count": 49,
"id": "398e8ff7",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/22-bsp-koch19.mp3',\n",
" '/mnt/data-ssd-1/data/podcasts/other/brain-science/audios/84-bsp-koch.mp3']"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"n = 5\n",
"l = [fp for fp in fps if repeat_guests[n] in fp.split(\"/\")[-1].lower()]\n",
"l"
]
},
{
"cell_type": "code",
"execution_count": 50,
"id": "1162a339",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"