Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -40,27 +40,41 @@ def evaluate_audio_similarity(original_audio_bytes, user_audio_bytes):
|
|
| 40 |
|
| 41 |
st.title("Audio Transcription and Similarity Checker")
|
| 42 |
|
| 43 |
-
#
|
| 44 |
-
st.
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
st.
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
|
|
|
|
|
|
|
|
|
| 59 |
st.session_state['initialized'] = True
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
|
| 65 |
if not original_audio_bytes:
|
| 66 |
original_audio = st.file_uploader("Or Upload Original Audio", type=["wav", "mp3"])
|
|
@@ -71,8 +85,19 @@ if input_method == "Record":
|
|
| 71 |
with st.spinner("Processing original audio..."):
|
| 72 |
st.audio(original_audio_bytes, format="audio/wav")
|
| 73 |
|
| 74 |
-
st.
|
| 75 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
if not user_audio_bytes:
|
| 78 |
user_audio = st.file_uploader("Or Upload User Audio", type=["wav", "mp3"])
|
|
@@ -96,7 +121,7 @@ if input_method == "Record":
|
|
| 96 |
st.write(f"**User Transcription:** {transcription_user}")
|
| 97 |
st.write(f"**Levenshtein Similarity Score:** {similarity_score:.2f}")
|
| 98 |
|
| 99 |
-
if similarity_score > 0.8:
|
| 100 |
st.success("The pronunciation is likely correct based on transcription similarity.")
|
| 101 |
else:
|
| 102 |
st.error("The pronunciation may be incorrect based on transcription similarity.")
|
|
|
|
| 40 |
|
| 41 |
st.title("Audio Transcription and Similarity Checker")
|
| 42 |
|
| 43 |
+
# Initialize the session state to control the view
|
| 44 |
+
if 'initialized' not in st.session_state:
|
| 45 |
+
st.session_state['initialized'] = False
|
| 46 |
+
|
| 47 |
+
# Button to initialize the recorders
|
| 48 |
+
if not st.session_state['initialized']:
|
| 49 |
+
|
| 50 |
+
st.write("Click the Loader below to initialize the audio recorders.")
|
| 51 |
+
|
| 52 |
+
if st.button('Click to start recording'):
|
| 53 |
+
init_button = audio_recorder(
|
| 54 |
+
text="",
|
| 55 |
+
recording_color="#e8b62c",
|
| 56 |
+
neutral_color="#6aa36f",
|
| 57 |
+
pause_threshold=0.1,
|
| 58 |
+
icon_name="", # You can change this to any Font Awesome solid icon
|
| 59 |
+
icon_size="4x",
|
| 60 |
+
auto_start=True
|
| 61 |
+
)
|
| 62 |
st.session_state['initialized'] = True
|
| 63 |
+
|
| 64 |
+
# If initialized, display the recorders
|
| 65 |
+
if st.session_state['initialized']:
|
| 66 |
+
|
| 67 |
+
st.subheader("Record or Upload Original Audio")
|
| 68 |
+
|
| 69 |
+
# Style the record button with the provided parameters
|
| 70 |
+
original_audio_bytes = audio_recorder(
|
| 71 |
+
text="Click to Record Audio",
|
| 72 |
+
recording_color="#e8b62c",
|
| 73 |
+
neutral_color="#6aa36f",
|
| 74 |
+
pause_threshold=30,
|
| 75 |
+
icon_name="microphone", # You can change this to any Font Awesome solid icon
|
| 76 |
+
icon_size="4x"
|
| 77 |
+
)
|
| 78 |
|
| 79 |
if not original_audio_bytes:
|
| 80 |
original_audio = st.file_uploader("Or Upload Original Audio", type=["wav", "mp3"])
|
|
|
|
| 85 |
with st.spinner("Processing original audio..."):
|
| 86 |
st.audio(original_audio_bytes, format="audio/wav")
|
| 87 |
|
| 88 |
+
st.subheader("Record or Upload User Audio")
|
| 89 |
+
|
| 90 |
+
st.write("")
|
| 91 |
+
|
| 92 |
+
# Style the user audio recorder similarly
|
| 93 |
+
user_audio_bytes = audio_recorder(
|
| 94 |
+
text="Click to Record Audio",
|
| 95 |
+
recording_color="#e86f6f",
|
| 96 |
+
neutral_color="#6a6faf",
|
| 97 |
+
pause_threshold=30,
|
| 98 |
+
icon_name="user", # You can change this to any Font Awesome solid icon
|
| 99 |
+
icon_size="4x"
|
| 100 |
+
)
|
| 101 |
|
| 102 |
if not user_audio_bytes:
|
| 103 |
user_audio = st.file_uploader("Or Upload User Audio", type=["wav", "mp3"])
|
|
|
|
| 121 |
st.write(f"**User Transcription:** {transcription_user}")
|
| 122 |
st.write(f"**Levenshtein Similarity Score:** {similarity_score:.2f}")
|
| 123 |
|
| 124 |
+
if similarity_score > 0.8: # Adjust the threshold as needed
|
| 125 |
st.success("The pronunciation is likely correct based on transcription similarity.")
|
| 126 |
else:
|
| 127 |
st.error("The pronunciation may be incorrect based on transcription similarity.")
|