Update app.py

app.py

@@ -1,30 +1,49 @@
+import torch
+
 import gradio as gr
+from transformers import pipeline
+from transformers.pipelines.audio_utils import ffmpeg_read
 
-def transcribe_speech(audio_file):
-    # Load the model
-    asr_model = gr.load("models/rngzhi/cs3264-project")
-    result = asr_model(audio_file)
-    return result['text']
+import tempfile
+import os
+
+MODEL_NAME = "rngzhi/cs3264-project"
+BATCH_SIZE = 8
+FILE_LIMIT_MB = 1000
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
 
-demo = gr.Blocks()
 
+def transcribe(inputs, task):
+    if inputs is None:
+        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
+
+    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
+    return  text
+
+
+demo = gr.Blocks()
 mic_transcribe = gr.Interface(
-    fn=transcribe_speech,
+    fn=transcribe,
     inputs=gr.Audio(sources="microphone", type="filepath"),
-    outputs=gr.Textbox(),
+    outputs="text",
 )
 
 file_transcribe = gr.Interface(
-    fn=transcribe_speech,
+    fn=transcribe,
     inputs=gr.Audio(sources="upload", type="filepath"),
-    outputs=gr.Textbox(),
+    outputs="text",
 )
 
-with demo:
-    gr.TabbedInterface(
-        [mic_transcribe, file_transcribe],
-        ["Transcribe Microphone", "Transcribe Audio File"],
-    )
 
+with demo:
+    gr.TabbedInterface([file_transcribe, mic_transcribe], ["Audio file", "Microphone"])
 
-demo.launch(share=True,auth=None,debug=False)
+demo.launch(debug=True)