working summarizer

openbb_terminal/forecast/whisper_view.py

@@ -15,6 +15,9 @@ from openbb_terminal.rich_config import console
 import warnings
 import os
 import whisper
+from transformers import pipeline
+
+
 from openbb_terminal.forecast.whisper_utils import (
     slugify,
     str2bool,
@@ -39,6 +42,7 @@ def get_audio(urls):
             "verbose": True,
             "format": "bestaudio",
             "outtmpl": os.path.join(temp_dir, "%(id)s.%(ext)s"),
+            "external_downloader_args": ["-loglevel", "panic"],
             "postprocessors": [
                 {
                     "preferredcodec": "mp3",
@@ -72,6 +76,9 @@ def transcribe_and_summarize(
     output_dir: str = "/Users/martinbufi/OpenBBTerminal/openbb_terminal/forecast/whisper_output",
 ):
 
+    # Use the pipeline to summarize the text
+    summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
+
     console.print("Transcribing and summarizing...")
     print(f"video: {video}")
     if model_name.endswith(".en"):
@@ -99,7 +106,18 @@ def transcribe_and_summarize(
         for segment in result["segments"]:
             all_text += segment["text"]
 
-        print(all_text)
+        original_text_length = len(all_text)
+
+        summary = summarizer(all_text)
+
+        summary_text = summary[0]["summary_text"]
+        summary_text_length = len(summary_text)
+        percent_reduction = round(
+            (1 - (summary_text_length / original_text_length)) * 100, 2
+        )
+
+        print(f"Summary (reduction {percent_reduction})")
+        print(summary_text)
 
         if subtitles_format == "vtt":
             vtt_path = os.path.join(output_dir, f"{slugify(title)}.vtt")