adds late night host app

late-night-host-app/README.md

@@ -0,0 +1,15 @@
+# Late Night Host App powered by Deepgram
+
+
+A test application used to download some YouTube videos from our favorite Late Night talk show hosts and transcribe them using Deepgram.
+
+
+## References
+* [Deepgram How to Video](https://www.youtube.com/watch?v=tt6fkobEtWk)
+* https://github.com/deepgram-devs/ExplainTuber-Analysis
+
+
+### Project Imports
+
+- I recommend using [yt-dlp](https://github.com/yt-dlp/yt-dlp) instead of [youtube-dl](https://github.com/ytdl-org/youtube-dl) due to this issue:
+  - https://github.com/ytdl-org/youtube-dl/issues/31530

late-night-host-app/deepgram_transcribe.py

@@ -0,0 +1,33 @@
+from deepgram import Deepgram as DG
+import asyncio, json, os
+
+# is there a way in python to get this into an environment variable?
+dg_key = "d222aa38a2749d3ba17340f08263acee4be1d4b8"
+dg = DG(dg_key)
+
+options = {
+    "diarize": True,
+    "punctuate": True,
+    "paragraphs": True,
+    "numerals": True,
+    "model": "general",
+    "tier": "enhanced",
+    "profanity_filter": True,
+}
+
+
+async def main():
+    videos = os.listdir("./audio")
+    for video in videos:
+        print("Currently processing:", video)
+        with open(f"./audio/{video}", "rb") as audio:
+            source = {"buffer": audio, "mimetype": "audio/mpeg3"}
+            res = await dg.transcription.prerecorded(source, options)
+
+            # outputs to a directory called transcripts
+            with open(f"./transcripts/{video[:-4]}.json", "w") as transcript:
+                json.dump(res, transcript)
+    return
+
+
+asyncio.run(main())

late-night-host-app/youtube_download.py

@@ -0,0 +1,37 @@
+import yt_dlp
+
+obrien_vids = [
+    "https://youtu.be/P4d8QrRJvsE",
+]
+
+kimmel_vids = [
+    "https://youtu.be/KA3KDbU5dqk",
+]
+
+meyers_vids = [
+    "https://youtu.be/q5UsZ2YdyKE",
+]
+
+colbert_vids = [
+    "https://youtu.be/PO508nFSIaM",
+]
+
+
+ydl_parameters = {
+    "format": "bestaudio/best",
+    "postprocessors": [
+        {
+            "key": "FFmpegExtractAudio",
+            "preferredcodec": "mp3",
+            "preferredquality": "192",
+        }
+    ],
+    # change this to change the path you want your downloads to be located
+    "outtmpl": "./audio/%(title)s.mp3",
+}
+with yt_dlp.YoutubeDL(ydl_parameters) as ydl:
+    ydl.download(obrien_vids)
+    ydl.download(kimmel_vids)
+    ydl.download(meyers_vids)
+    ydl.download(colbert_vids)
+    print()