Update app.py

app.py

@@ -1,11 +1,9 @@
 import spaces
 import torch
-
 import gradio as gr
 import yt_dlp as youtube_dl
 from transformers import pipeline
 from transformers.pipelines.audio_utils import ffmpeg_read
-
 import tempfile
 import os
 
@@ -23,7 +21,6 @@ pipe = pipeline(
     device=device,
 )
 
-
 @spaces.GPU
 def transcribe(inputs, task):
     if inputs is None:
@@ -37,10 +34,48 @@ def transcribe(inputs, task):
     word_timestamps = []
     for chunk in timestamps:
         for word_info in chunk["words"]:
-            word_timestamps.append(f"{word_info['word']} [{word_info['start']}-{word_info['end']}]")
+            word_timestamps.append(f"{word_info['word']} [{word_info['start']:.2f}-{word_info['end']:.2f}]")
     
     return "\n".join(word_timestamps)
 
+def _return_yt_html_embed(yt_url):
+    video_id = yt_url.split("?v=")[-1]
+    HTML_str = (
+        f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
+        " </center>"
+    )
+    return HTML_str
+
+def download_yt_audio(yt_url, filename):
+    info_loader = youtube_dl.YoutubeDL()
+    
+    try:
+        info = info_loader.extract_info(yt_url, download=False)
+    except youtube_dl.utils.DownloadError as err:
+        raise gr.Error(str(err))
+    
+    file_length = info["duration_string"]
+    file_h_m_s = file_length.split(":")
+    file_h_m_s = [int(sub_length) for sub_length in file_h_m_s]
+    
+    if len(file_h_m_s) == 1:
+        file_h_m_s.insert(0, 0)
+    if len(file_h_m_s) == 2:
+        file_h_m_s.insert(0, 0)
+    file_length_s = file_h_m_s[0] * 3600 + file_h_m_s[1] * 60 + file_h_m_s[2]
+    
+    if file_length_s > YT_LENGTH_LIMIT_S:
+        yt_length_limit_hms = time.strftime("%HH:%MM:%SS", time.gmtime(YT_LENGTH_LIMIT_S))
+        file_length_hms = time.strftime("%HH:%MM:%SS", time.gmtime(file_length_s))
+        raise gr.Error(f"Maximum YouTube length is {yt_length_limit_hms}, got {file_length_hms} YouTube video.")
+    
+    ydl_opts = {"outtmpl": filename, "format": "worstvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best"}
+    
+    with youtube_dl.YoutubeDL(ydl_opts) as ydl:
+        try:
+            ydl.download([yt_url])
+        except youtube_dl.utils.ExtractorError as err:
+            raise gr.Error(str(err))
 
 @spaces.GPU
 def yt_transcribe(yt_url, task, max_filesize=75.0):
@@ -63,12 +98,11 @@ def yt_transcribe(yt_url, task, max_filesize=75.0):
     word_timestamps = []
     for chunk in timestamps:
         for word_info in chunk["words"]:
-            word_timestamps.append(f"{word_info['word']} [{word_info['start']}-{word_info['end']}]")
+            word_timestamps.append(f"{word_info['word']} [{word_info['start']:.2f}-{word_info['end']:.2f}]")
     
     return html_embed_str, "\n".join(word_timestamps)
 
 
-
 demo = gr.Blocks()
 
 mf_transcribe = gr.Interface(
@@ -77,7 +111,7 @@ mf_transcribe = gr.Interface(
         gr.Audio(sources="microphone", type="filepath"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
     ],
-    outputs=["text", "text"],  # Output both text and timestamps
+    outputs="text",
     title="Whisper Large V3: Transcribe Audio",
     description=(
         "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
@@ -93,7 +127,7 @@ file_transcribe = gr.Interface(
         gr.Audio(sources="upload", type="filepath", label="Audio file"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
     ],
-    outputs=["text", "text"],  # Output both text and timestamps
+    outputs="text",
     title="Whisper Large V3: Transcribe Audio",
     description=(
         "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
@@ -109,7 +143,7 @@ yt_transcribe = gr.Interface(
         gr.Textbox(lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe")
     ],
-    outputs=["html", "text", "text"],  # Output both text and timestamps
+    outputs=["html", "text"],
     title="Whisper Large V3: Transcribe YouTube",
     description=(
         "Transcribe long-form YouTube videos with the click of a button! Demo uses the checkpoint"