add openai version

app.py

@@ -1 +1 @@
-run_demo_multi_models.py
+run_demo_openai.py

packages.txt

@@ -1 +1,2 @@
 ffmpeg
+git+https://github.com/openai/whisper.git 

run_demo_openai.py

@@ -0,0 +1,132 @@
+import logging
+import warnings
+
+import gradio as gr
+import pytube as pt
+import torch
+from huggingface_hub import hf_hub_download, model_info
+from transformers.utils.logging import disable_progress_bar
+import whisper
+
+
+warnings.filterwarnings("ignore")
+disable_progress_bar()
+
+MODEL_NAME = "bofenghuang/whisper-large-v2-cv11-french"
+CHECKPOINT_FILENAME = "checkpoint_openai.pt"
+
+logging.basicConfig(
+    format="%(asctime)s [%(levelname)s] [%(name)s] %(message)s",
+    datefmt="%Y-%m-%dT%H:%M:%SZ",
+)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.DEBUG)
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+downloaded_model_path = hf_hub_download(repo_id=MODEL_NAME, filename=CHECKPOINT_FILENAME)
+
+model = whisper.load_model(downloaded_model_path, device=device)
+logger.info(f"Model has been loaded on device `{device}`")
+
+gen_kwargs = {
+    "task": "transcribe",
+    "language": "fr",
+    # "without_timestamps": True,
+    # decode options
+    # "beam_size": 5,
+    # "patience": 2,
+    # disable fallback
+    # "compression_ratio_threshold": None,
+    # "logprob_threshold": None,
+    # vad threshold
+    # "no_speech_threshold": None,
+}
+
+def transcribe(microphone, file_upload):
+    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
+
+    text = model.transcribe(file, **gen_kwargs)["text"]
+
+    logger.info(f"Transcription: {text}")
+
+    return warn_output + text
+
+
+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):
+    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 = model.transcribe("audio.mp3", **gen_kwargs)["text"]
+
+    logger.info(f'Transcription of "{yt_url}": {text}')
+
+    return html_embed_str, text
+
+
+demo = gr.Blocks()
+
+mf_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.inputs.Audio(source="microphone", type="filepath", optional=True, label="Record"),
+        gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Upload File"),
+    ],
+    # outputs="text",
+    outputs=gr.outputs.Textbox(label="Transcription"),
+    layout="horizontal",
+    theme="huggingface",
+    title="Whisper French Demo 🇫🇷 : Transcribe Audio",
+    description=(
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the the fine-tuned"
+        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")],
+    # outputs=["html", "text"],
+    outputs=[
+        gr.outputs.HTML(label="YouTube Page"),
+        gr.outputs.Textbox(label="Transcription"),
+    ],
+    layout="horizontal",
+    theme="huggingface",
+    title="Whisper French Demo 🇫🇷 : Transcribe YouTube",
+    description=(
+        "Transcribe long-form YouTube videos with the click of a button! Demo uses the the fine-tuned checkpoint:"
+        f" [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files of"
+        " arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface([mf_transcribe, yt_transcribe], ["Transcribe Audio", "Transcribe YouTube"])
+
+# demo.launch(server_name="0.0.0.0", debug=True, share=True)
+demo.launch(enable_queue=True)