Update app.py

app.py

@@ -3,12 +3,12 @@ import torch
 import gradio as gr
 import pytube as pt
 from transformers import pipeline
-from huggingface_hub import model_info
 
-MODEL_NAME = "openai/whisper-medium" #this always needs to stay in line 8 :D sorry for the hackiness
-lang = "en"
+MODEL_NAME = "openai/whisper-medium"
+BATCH_SIZE = 8
 
 device = 0 if torch.cuda.is_available() else "cpu"
+
 pipe = pipeline(
     task="automatic-speech-recognition",
     model=MODEL_NAME,
@@ -16,9 +16,13 @@ pipe = pipeline(
     device=device,
 )
 
-pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language=lang, task="transcribe")
 
-def transcribe(microphone, file_upload):
+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 = (
@@ -31,7 +35,9 @@ def transcribe(microphone, file_upload):
 
     file = microphone if microphone is not None else file_upload
 
-    text = pipe(file)["text"]
+    pipe.model.config.forced_decoder_ids = [[2, transcribe_token_id if task=="transcribe" else translate_token_id]]
+
+    text = pipe(file, batch_size=BATCH_SIZE)["text"]
 
     return warn_output + text
 
@@ -45,13 +51,15 @@ def _return_yt_html_embed(yt_url):
     return HTML_str
 
 
-def yt_transcribe(yt_url):
+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")
 
-    text = pipe("audio.mp3")["text"]
+    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
 
@@ -63,26 +71,34 @@ mf_transcribe = gr.Interface(
     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",
     layout="horizontal",
     theme="huggingface",
-    title="Speech to Text using Open AI",
+    title="Whisper Large V2: Transcribe Audio",
     description=(
-        "Transcribe long-form microphone or audio inputs with the click of a button!"
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
+        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
+        " of arbitrary length."
     ),
     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")],
+    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 using Open AI Transcribe YouTube",
+    title="Whisper Large V2: Transcribe YouTube",
     description=(
-        "Transcribe long-form YouTube videos with the click of a button! "
+        "Transcribe long-form YouTube videos with the click of a button! Demo uses the checkpoint"
+        f" [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe video files of"
+        " arbitrary length."
     ),
     allow_flagging="never",
 )
@@ -90,4 +106,4 @@ yt_transcribe = gr.Interface(
 with demo:
     gr.TabbedInterface([mf_transcribe, yt_transcribe], ["Transcribe Audio", "Transcribe YouTube"])
 
-demo.launch(enable_queue=True)
+demo.launch(enable_queue=True)