Update app.py

app.py

@@ -1,43 +1,43 @@
 import streamlit as st
 import speech_recognition as sr
-from pydub import AudioSegment
 
-# Title and description
-st.title("Audio Trimming App")
-st.write("Upload an audio file and enter the desired text. The app will trim the audio to match the specified text.")
-
-# Upload an audio file
-uploaded_audio = st.file_uploader("Upload an audio file", type=["wav", "mp3"])
-
-# Input text box
-input_text = st.text_area("Enter the desired text")
-
-# Function to trim audio
-def trim_audio(input_audio, input_text):
-    if not input_audio:
-        return
-
-    # Load the audio file and recognize text
+def transcribe_audio(audio_file):
     recognizer = sr.Recognizer()
-    with sr.AudioFile(input_audio) as source:
-        audio = recognizer.record(source)
-        audio_text = recognizer.recognize_google(audio)
 
-    # Process the audio and create a trimmed audio
-    trimmed_audio = AudioSegment.silent(duration=0)
-    text_to_match = input_text.lower()
-    audio_text = audio_text.lower()
-
-    for word in audio_text.split():
-        if word in text_to_match:
-            trimmed_audio += AudioSegment.from_file(input_audio)
-
-    # Export the trimmed audio to a file
-    trimmed_audio.export("trimmed_audio.mp3", format="mp3")
-
-    # Display the trimmed audio
-    st.audio("trimmed_audio.mp3", format="audio/mp3")
+    with sr.AudioFile(audio_file) as source:
+        audio = recognizer.record(source)
 
-# Process button
-if st.button("Process"):
-    trim_audio(uploaded_audio, input_text)
+    try:
+        transcription = recognizer.recognize_google(audio)
+        return transcription
+    except sr.UnknownValueError:
+        return ""
+
+def filter_words(input_text, transcribed_text):
+    transcribed_words = transcribed_text.split()
+    filtered_text = input_text
+    for word in transcribed_words:
+        filtered_text = filtered_text.replace(word, "")
+    return filtered_text
+
+st.title("Voice Cloning Word Filter")
+
+uploaded_audio = st.file_uploader("Upload an audio file", type=["wav", "mp3", "ogg"])
+input_text = st.text_area("Enter input text")
+
+if uploaded_audio is not None:
+    st.audio(uploaded_audio, format="audio/wav")
+
+if st.button("Transcribe and Filter"):
+    if uploaded_audio is not None:
+        transcribed_text = transcribe_audio(uploaded_audio)
+        filtered_text = filter_words(input_text, transcribed_text)
+        st.subheader("Transcribed Text:")
+        st.write(transcribed_text)
+        st.subheader("Filtered Output Text:")
+        st.write(filtered_text)
+
+st.write(
+    "Note: This is a simple demonstration of filtering words from an audio transcript. The accuracy of word filtering "
+    "depends on the quality of the audio and the performance of the ASR engine."
+)