... not done...

core/text_to_speech.py

@@ -39,7 +39,7 @@ class openai_tts(object):
 
     def set_tts_speed(self, speed):
         if not (0.25 <= speed <= 4.0):
-            raise ValueError(f"La vitesse doit être comprise entre 0.25 et 4.0. Valeur fournie : {speed}")
+            raise ValueError(f"[TTS] - Speed must be between 0.25 and 4.0. Provided value: {speed}")
         else:
             self.speed = speed
         return self
@@ -47,14 +47,14 @@ class openai_tts(object):
     def set_tts_voice(self, voice):
         voix_valides = ["alloy", "echo", "fable", "onyx", "nova", "shimmer"]
         if voice not in voix_valides:
-            raise ValueError(f"Invalid TTS voice: {voice}. Valid voices are: {', '.join(voix_valides)}.")
+            raise ValueError(f"[TTS] - Invalid TTS voice: {voice}. Valid voices are: {', '.join(voix_valides)}.")
         else:
             self.tts_voice = voice
         return self
 
     def set_tts_model(self, model):
         if model not in ["tts-1", "tts-1-hd"]:
-            raise ValueError(f"Invalid TTS model: {model}. Valid models are 'tts-1' and 'tts-1-hd'.")
+            raise ValueError(f"[TTS] - Invalid TTS model: {model}. Valid models are 'tts-1' and 'tts-1-hd'.")
         else:
             self.tts_model = model
         return self
@@ -65,7 +65,7 @@ class openai_tts(object):
 
     def set_response_format(self, format: str):
         if format not in self.supported_formats:
-            raise ValueError(f"Unsupported format: {format}. Supported formats are: {', '.join(self.supported_formats)}")
+            raise ValueError(f"[TTS] - Unsupported format: {format}. Supported formats are: {', '.join(self.supported_formats)}")
         else:
             self.response_format = format
         return self
@@ -85,11 +85,13 @@ class openai_tts(object):
             input_text (str): Le texte à convertir en parole.
 
         Returns:
-            Tuple[Optional[bytes], float]: Un tuple contenant les octets audio et la durée de l'audio en secondes.
+            Dict[str, Union[float, str]]: Un dictionnaire contenant:
+                - 'audio_duration' (float): La durée de l'audio en secondes.
+                - 'data_bytes' (str): Les données audio encodées en base64.
         """
         response = self.client.audio.speech.create(
             model=self.tts_model,
-            voice=self.tts_voice ,#st.session_state.tts_voice,
+            voice=self.tts_voice,
             input=input_text,
             response_format=self.response_format,
             speed=self.speed