up

run_demo_ct2.py

@@ -114,6 +114,7 @@ def maybe_load_cached_pipeline(model_name):
         # downloaded_model_path = hf_hub_download(repo_id=model_name, filename=CHECKPOINT_FILENAME)
         # downloaded_model_path = snapshot_download(repo_id=model_name)
         downloaded_model_path = snapshot_download(repo_id=model_name, allow_patterns="ctranslate2/*")
+        downloaded_model_path = f"{downloaded_model_path}/ctranslate2"
 
         # model = whisper.load_model(downloaded_model_path, device=device)
         model = WhisperModel(downloaded_model_path, device=device, compute_type="float16")

run_demo_hf.py

@@ -11,7 +11,11 @@ from transformers.utils.logging import disable_progress_bar
 warnings.filterwarnings("ignore")
 disable_progress_bar()
 
-MODEL_NAME = "bofenghuang/whisper-large-v2-cv11-french"
+# MODEL_NAME = "bofenghuang/whisper-large-v2-cv11-french"
+MODEL_NAME = "bofenghuang/whisper-large-v3-french"
+# MODEL_NAME = "/home/bhuang/transformers/examples/pytorch/speech-recognition/outputs/hf_whisper/whisper-large-v3-ft-french-pnc-ep5-bs280-lr4e6-wd001-audioaug-specaug"
+# MODEL_NAME = "/home/bhuang/transformers/examples/pytorch/speech-recognition/outputs/hf_whisper/tmp_model"
+# MODEL_NAME = "/projects/bhuang/models/asr/public/whisper-large-v3-french"
 CHUNK_LENGTH_S = 30
 
 logging.basicConfig(
@@ -80,38 +84,38 @@ 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"),
+        gr.components.Audio(sources="microphone", type="filepath", label="Record"),
+        gr.components.Audio(sources="upload", type="filepath", label="Upload File"),
     ],
     # outputs="text",
-    outputs=gr.outputs.Textbox(label="Transcription"),
-    layout="horizontal",
+    outputs=gr.components.Textbox(label="Transcription", show_copy_button=True),
+    # 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."
-    ),
+    # 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")],
+    inputs=[gr.components.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"),
+        gr.components.HTML(label="YouTube Page"),
+        gr.components.Textbox(label="Transcription", show_copy_button=True),
     ],
-    layout="horizontal",
+    # 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."
-    ),
+    # 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",
 )
 
@@ -119,4 +123,6 @@ 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)
+# demo.launch(enable_queue=True)
+# see https://github.com/gradio-app/gradio/issues/2551
+demo.queue(max_size=10).launch(server_name="0.0.0.0", debug=True, share=True, ssl_certfile="/home/bhuang/tools/cert.pem", ssl_keyfile="/home/bhuang/tools/key.pem", ssl_verify=False)