File size: 4,316 Bytes
3282984 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 |
import whisper
import os
import streamlit as st
import time
@st.cache_resource
def load_whisper_model(model_name="base"):
# Create a placeholder for notifications
notify = st.empty()
notify.info(f"Loading Whisper '{model_name}' model for transcription... This may take a moment.", icon="⏳")
whisper_model = whisper.load_model(model_name)
notify.success(f"Whisper '{model_name}' model loaded successfully!")
time.sleep(2) # Show success for 2 seconds
notify.empty() # Clear the notification
return whisper_model
whisper_model_instance = load_whisper_model("base")
def transcribe_audio(audio_path):
try:
result = whisper_model_instance.transcribe(audio_path)
return result["text"]
except Exception as e:
st.error(f"Error during Whisper transcription: {e}")
return "" # Return empty string if transcription fails
# @st.cache_resource
# def load_whisper_model(model_name="base"):
# status_placeholder = st.empty()
# with status_placeholder.status(f"Loading Whisper '{model_name}' model...", expanded=True) as status:
# st.write("Initializing model download and setup...")
# try:
# whisper_model = whisper.load_model(model_name)
# status.update(label=f"Whisper '{model_name}' model loaded successfully!", state="complete", expanded=False)
# # st.session_state is available even in cached functions if they are part of the app's scope
# st.session_state['whisper_model_loaded'] = True
# return whisper_model
# except Exception as e:
# status.update(label=f"Error loading Whisper '{model_name}' model: {e}", state="error")
# st.error("Model loading failed! Please check your internet connection or try again.")
# st.session_state['whisper_model_loaded'] = False # Mark as failed
# raise # Re-raise the exception to propagate it
# # Load the desired Whisper model globally within this module
# whisper_model_instance = load_whisper_model("base")
# def transcribe_audio(audio_path):
# """
# Transcribes the audio file.
# """
# try:
# result = whisper_model_instance.transcribe(audio_path)
# return result["text"]
# except Exception as e:
# st.error(f"Error during Whisper transcription: {e}")
# return ""
# @st.cache_resource
# def load_whisper_model(model_name="base"):
# # st.info(f"Loading Whisper '{model_name}' model for transcription... This may take a moment.", icon="⏳")
# whisper_model = whisper.load_model(model_name)
# # st.success(f"Whisper '{model_name}' model loaded successfully!")
# return whisper_model
# whisper_model_instance = load_whisper_model("base")
# def transcribe_audio(audio_path):
# try:
# result = whisper_model_instance.transcribe(audio_path)
# return result["text"]
# except Exception as e:
# st.error(f"Error during Whisper transcription: {e}")
# return "" # Return empty string if transcription fails
# @st.cache_resource
# def load_whisper_model(model_name="base"):
# with st.status(f"Loading Whisper '{model_name}' model...", expanded=True) as status:
# st.write("Initializing model download and setup...")
# try:
# whisper_model = whisper.load_model(model_name)
# # Update the status once complete
# status.update(label=f"Whisper '{model_name}' model loaded successfully!", state="complete", expanded=False)
# return whisper_model
# except Exception as e:
# # Update the status on error
# status.update(label=f"Error loading Whisper '{model_name}' model: {e}", state="error")
# st.error("Model loading failed! Please check your internet connection or try again.")
# raise # Re-raise the exception to propagate it
# # Load the desired Whisper model globally within this module
# whisper_model_instance = load_whisper_model("base")
# def transcribe_audio(audio_path):
# try:
# result = whisper_model_instance.transcribe(audio_path)
# return result["text"]
# except Exception as e:
# st.error(f"Error during Whisper transcription: {e}")
# return "" |