Spaces:
Running
Running
drewThomasson
commited on
Commit
β’
a553ade
1
Parent(s):
5dc8e6c
Update app.py
Browse files
app.py
CHANGED
@@ -93,7 +93,7 @@ def create_speaker_with_transcription(audio_file):
|
|
93 |
audio_file (file): Uploaded reference audio file.
|
94 |
|
95 |
Returns:
|
96 |
-
dict
|
97 |
"""
|
98 |
logger.info("Received Voice Cloning request with audio file.")
|
99 |
|
@@ -110,7 +110,7 @@ def create_speaker_with_transcription(audio_file):
|
|
110 |
if not transcript.strip():
|
111 |
logger.error("Transcription resulted in empty text.")
|
112 |
os.remove(temp_audio_path)
|
113 |
-
return
|
114 |
|
115 |
# Create speaker using the transcribed text
|
116 |
speaker = interface.create_speaker(temp_audio_path, transcript)
|
@@ -123,7 +123,7 @@ def create_speaker_with_transcription(audio_file):
|
|
123 |
return speaker
|
124 |
except Exception as e:
|
125 |
logger.error(f"Error during speaker creation: {e}")
|
126 |
-
return
|
127 |
|
128 |
def generate_tts_cloned(text, temperature, repetition_penalty, max_length, speaker):
|
129 |
"""
|
@@ -143,7 +143,7 @@ def generate_tts_cloned(text, temperature, repetition_penalty, max_length, speak
|
|
143 |
logger.info(f"Parameters - Text: {text}, Temperature: {temperature}, Repetition Penalty: {repetition_penalty}, Max Length: {max_length}, Speaker Provided: {speaker is not None}")
|
144 |
|
145 |
try:
|
146 |
-
if speaker
|
147 |
logger.error("Speaker configuration is missing.")
|
148 |
return None
|
149 |
|
@@ -230,12 +230,12 @@ with gr.Blocks() as demo:
|
|
230 |
with gr.Row():
|
231 |
reference_audio = gr.Audio(
|
232 |
label="π Reference Audio",
|
233 |
-
type="filepath"
|
234 |
)
|
235 |
|
236 |
create_speaker_button = gr.Button("π€ Create Speaker")
|
237 |
|
238 |
-
speaker_info = gr.JSON(label="ποΈ Speaker Configuration"
|
239 |
|
240 |
with gr.Row():
|
241 |
generate_cloned_speech = gr.Textbox(
|
|
|
93 |
audio_file (file): Uploaded reference audio file.
|
94 |
|
95 |
Returns:
|
96 |
+
dict: Speaker configuration or empty dict if failed.
|
97 |
"""
|
98 |
logger.info("Received Voice Cloning request with audio file.")
|
99 |
|
|
|
110 |
if not transcript.strip():
|
111 |
logger.error("Transcription resulted in empty text.")
|
112 |
os.remove(temp_audio_path)
|
113 |
+
return {}
|
114 |
|
115 |
# Create speaker using the transcribed text
|
116 |
speaker = interface.create_speaker(temp_audio_path, transcript)
|
|
|
123 |
return speaker
|
124 |
except Exception as e:
|
125 |
logger.error(f"Error during speaker creation: {e}")
|
126 |
+
return {}
|
127 |
|
128 |
def generate_tts_cloned(text, temperature, repetition_penalty, max_length, speaker):
|
129 |
"""
|
|
|
143 |
logger.info(f"Parameters - Text: {text}, Temperature: {temperature}, Repetition Penalty: {repetition_penalty}, Max Length: {max_length}, Speaker Provided: {speaker is not None}")
|
144 |
|
145 |
try:
|
146 |
+
if not speaker:
|
147 |
logger.error("Speaker configuration is missing.")
|
148 |
return None
|
149 |
|
|
|
230 |
with gr.Row():
|
231 |
reference_audio = gr.Audio(
|
232 |
label="π Reference Audio",
|
233 |
+
type="filepath" # Receive the path to the uploaded file
|
234 |
)
|
235 |
|
236 |
create_speaker_button = gr.Button("π€ Create Speaker")
|
237 |
|
238 |
+
speaker_info = gr.JSON(label="ποΈ Speaker Configuration") # Removed interactive=False
|
239 |
|
240 |
with gr.Row():
|
241 |
generate_cloned_speech = gr.Textbox(
|