Tonic commited on
Commit
603805b
1 Parent(s): 6ddb578

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +19 -23
app.py CHANGED
@@ -199,41 +199,37 @@ def convert_text_to_speech(input_text: str, source_language: str, target_languag
199
  client = Client("https://facebook-seamless-m4t.hf.space/--replicas/8cllp/")
200
 
201
  try:
 
202
  result = client.predict(
203
- "T2ST",
204
- "text",
205
- None,
206
- None,
207
- input_text,
208
- source_language,
209
- target_language,
210
- api_name="/run",
211
  )
212
-
213
  # Initialize variables
214
  translated_text = ""
215
  audio_file_path = ""
216
 
217
- # Process each item in the result
218
- for item in result:
219
- if isinstance(item, str):
220
- # Check if the item is likely a URL
221
- if item.startswith('http://') or item.startswith('https://'):
222
- continue
223
- # Assign the first non-URL string as the translated text
224
- if not translated_text:
225
- translated_text = item
226
- elif isinstance(item, tuple) and len(item) == 2:
227
- # Assuming the item is a tuple containing sample rate and audio data
228
- audio_file_path = save_audio(item) # Save the audio file
229
- break
230
 
231
  return audio_file_path, translated_text
232
 
233
  except Exception as e:
234
  return None, f"Error in text-to-speech conversion: {str(e)}"
235
 
236
-
237
  def process_image(image_input):
238
  # Initialize the Gradio client with the URL of the Gradio server
239
  client = Client("https://adept-fuyu-8b-demo.hf.space/--replicas/pqjvl/")
 
199
  client = Client("https://facebook-seamless-m4t.hf.space/--replicas/8cllp/")
200
 
201
  try:
202
+ # Make a prediction request to the client
203
  result = client.predict(
204
+ task_name="T2ST",
205
+ audio_source="text", # Since we are doing text-to-speech
206
+ input_audio_mic=None,
207
+ input_audio_file=None,
208
+ input_text=input_text,
209
+ source_language=source_language,
210
+ target_language=target_language,
211
+ api_name="/run"
212
  )
213
+
214
  # Initialize variables
215
  translated_text = ""
216
  audio_file_path = ""
217
 
218
+ # Process the result
219
+ if result:
220
+ # Assuming the result is a tuple containing audio data and translated text
221
+ audio_data, translated_text = result
222
+
223
+ if audio_data:
224
+ # Assuming audio_data is a tuple containing sample rate and audio numpy array
225
+ sample_rate, audio_numpy_array = audio_data
226
+ audio_file_path = save_audio(sample_rate, audio_numpy_array) # Save the audio file
 
 
 
 
227
 
228
  return audio_file_path, translated_text
229
 
230
  except Exception as e:
231
  return None, f"Error in text-to-speech conversion: {str(e)}"
232
 
 
233
  def process_image(image_input):
234
  # Initialize the Gradio client with the URL of the Gradio server
235
  client = Client("https://adept-fuyu-8b-demo.hf.space/--replicas/pqjvl/")