Spaces:
Sleeping
Sleeping
muzammil-eds
commited on
Commit
•
2910afc
1
Parent(s):
be99e76
Update app.py
Browse files
app.py
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
import streamlit as st
|
2 |
import requests
|
3 |
import Levenshtein
|
|
|
4 |
from io import BytesIO
|
5 |
import librosa
|
6 |
import torch
|
@@ -64,44 +65,15 @@ st.title("Audio Transcription and Similarity Checker")
|
|
64 |
|
65 |
# Choose between upload or record
|
66 |
st.sidebar.header("Input Method")
|
67 |
-
input_method = st.sidebar.selectbox("Choose Input Method", ["
|
68 |
|
69 |
original_audio_bytes = None
|
70 |
user_audio_bytes = None
|
71 |
|
72 |
-
if input_method == "Upload":
|
73 |
-
# Upload original audio file
|
74 |
-
original_audio = st.file_uploader("Upload Original Audio", type=["wav", "mp3"])
|
75 |
-
# Upload user audio file
|
76 |
-
user_audio = st.file_uploader("Upload User Audio", type=["wav", "mp3"])
|
77 |
|
78 |
-
|
79 |
-
original_audio_bytes = original_audio.read()
|
80 |
-
st.audio(original_audio_bytes, format="audio/wav")
|
81 |
-
if user_audio:
|
82 |
-
user_audio_bytes = user_audio.read()
|
83 |
-
st.audio(user_audio_bytes, format="audio/wav")
|
84 |
-
|
85 |
-
# Add a button to perform the test
|
86 |
-
if original_audio_bytes and user_audio_bytes:
|
87 |
-
if st.button("Perform Testing"):
|
88 |
-
with st.spinner("Performing transcription and similarity testing..."):
|
89 |
-
transcription_original, transcription_user, similarity_score = evaluate_audio_similarity(original_audio_bytes, user_audio_bytes)
|
90 |
-
|
91 |
-
# Display results
|
92 |
-
st.markdown("---")
|
93 |
-
st.subheader("Transcriptions and Similarity Score")
|
94 |
-
st.write(f"**Original Transcription:** {transcription_original}")
|
95 |
-
st.write(f"**User Transcription:** {transcription_user}")
|
96 |
-
st.write(f"**Levenshtein Similarity Score:** {similarity_score:.2f}")
|
97 |
-
|
98 |
-
if similarity_score > 0.8: # Adjust the threshold as needed
|
99 |
-
st.success("The pronunciation is likely correct based on transcription similarity.")
|
100 |
-
else:
|
101 |
-
st.error("The pronunciation may be incorrect based on transcription similarity.")
|
102 |
-
|
103 |
-
elif input_method == "Record":
|
104 |
st.write("Record or Upload Original Audio")
|
|
|
105 |
original_audio_bytes = audio_recorder(key="original_audio_recorder", pause_threshold=30, icon_size='4x')
|
106 |
|
107 |
if not original_audio_bytes:
|
|
|
1 |
import streamlit as st
|
2 |
import requests
|
3 |
import Levenshtein
|
4 |
+
import time
|
5 |
from io import BytesIO
|
6 |
import librosa
|
7 |
import torch
|
|
|
65 |
|
66 |
# Choose between upload or record
|
67 |
st.sidebar.header("Input Method")
|
68 |
+
input_method = st.sidebar.selectbox("Choose Input Method", ["Record"])
|
69 |
|
70 |
original_audio_bytes = None
|
71 |
user_audio_bytes = None
|
72 |
|
|
|
|
|
|
|
|
|
|
|
73 |
|
74 |
+
if input_method == "Record":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
75 |
st.write("Record or Upload Original Audio")
|
76 |
+
time.sleep(2)
|
77 |
original_audio_bytes = audio_recorder(key="original_audio_recorder", pause_threshold=30, icon_size='4x')
|
78 |
|
79 |
if not original_audio_bytes:
|