Create app.py

app.py

@@ -0,0 +1,107 @@
+import torch
+
+import gradio as gr
+import pytube as pt
+from transformers import pipeline
+
+MODEL_NAME = "openai/whisper-large-v2"
+BATCH_SIZE = 8
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
+
+
+all_special_ids = pipe.tokenizer.all_special_ids
+transcribe_token_id = all_special_ids[-5]
+translate_token_id = all_special_ids[-6]
+
+
+def transcribe(microphone, file_upload, task):
+    warn_output = ""
+    if (microphone is not None) and (file_upload is not None):
+        warn_output = (
+            "WARNING: You've uploaded an audio file and used the microphone. "
+            "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
+        )
+
+    elif (microphone is None) and (file_upload is None):
+        return "ERROR: You have to either use the microphone or upload an audio file"
+
+    file = microphone if microphone is not None else file_upload
+
+    pipe.model.config.forced_decoder_ids = [[2, transcribe_token_id if task=="transcribe" else translate_token_id]]
+
+    textt = pipe(file, batch_size=BATCH_SIZE)["text"]
+
+    with open('outt.txt', 'a+') as sw:
+        sw.writelines(textt)
+
+    return [textt,"outt.txt"]
+
+
+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 yt_transcribe(yt_url, task):
+    yt = pt.YouTube(yt_url)
+    html_embed_str = _return_yt_html_embed(yt_url)
+    stream = yt.streams.filter(only_audio=True)[0]
+    stream.download(filename="audio.mp3")
+
+    pipe.model.config.forced_decoder_ids = [[2, transcribe_token_id if task=="transcribe" else translate_token_id]]
+
+    text = pipe("audio.mp3", batch_size=BATCH_SIZE)["text"]
+
+    return html_embed_str, text
+
+demo = gr.Blocks()
+output_2 = gr.File(label="Download")
+description = """This application displays transcribed text for given audio input <img src="https://i.ibb.co/J5DscKw/GVP-Womens.jpg" width=100px>"""
+mf_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.inputs.Audio(source="microphone", type="filepath", optional=True),
+        gr.inputs.Audio(source="upload", type="filepath", optional=True),
+        gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
+    ],
+    outputs=["text",output_2],
+    layout="horizontal",
+    theme="huggingface",
+    title="Speech to Text Converter using OpenAI Whisper Model",
+    description= description,
+    allow_flagging="never",
+)
+
+yt_transcribe = gr.Interface(
+    fn=yt_transcribe,
+    inputs=[
+        gr.inputs.Textbox(lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
+        gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe")
+    ],
+    outputs=["html", "text"],
+    layout="horizontal",
+    theme="huggingface",
+    title="Speech to Text Converter using OpenAI Whisper Model",
+    description=(
+        "Transcribe YouTube Videos to Text"
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface([mf_transcribe, yt_transcribe], ["Transcribe Audio", "Transcribe YouTube"])
+
+demo.launch(enable_queue=True)