✨ make final fixes

src/api/main.py

@@ -1,4 +1,5 @@
 from typing import TypeAlias, Tuple, Any
+from pathlib import Path
 import logging
 import hashlib
 import flask
@@ -15,6 +16,8 @@
 
 Response: TypeAlias = Tuple[dict, int] # Just a type alias for the response
 
+DEFAULT_PATH = 'lele.wav'
+
 def run_api(_voice_cloner: VoiceCloner) -> None:
     """
     Runs the API for serving feature extraction method
@@ -31,10 +34,6 @@ def jsonify_error(err: errs.ErrorResponse) -> Response:
         json_err, status_code = err
         return flask.jsonify(json_err), status_code
 
-    # Returning an error if got a wrong method
-    if flask.request.method != "POST":
-        return jsonify_error(errs.INVALID_METHOD)
-
     # Asserting that the request has the necessary data
     def validate_request() -> bool:
         if "data" not in flask.request.json:
@@ -55,14 +54,14 @@ def validate_request() -> bool:
 
     # Cloning the voice
     try:
-        voice = voice_cloner.clone_voice(text)
+        voice = voice_cloner.save_voice_locally(text, Path(DEFAULT_PATH))
     except Exception as exception:
         logging.error(f"Failed to process the image: {exception}")
         return jsonify_error(errs.INTERNAL_ERROR)
 
     # Updating that the claim was submitted
     try:
-        return flask.send_file(voice, as_attachment=True, mimetype='.wav', download_name='lele.wav')
+        return flask.send_file(Path('../../' + DEFAULT_PATH), as_attachment=True, mimetype='audio/wav', download_name='lele.wav')
     except Exception as exception:
         logging.error(f"Failed while finalizing the request: {exception}")
         return jsonify_error(errs.INTERNAL_ERROR)

src/cloner/main.py

@@ -4,6 +4,7 @@
 from pathlib import Path
 from typing import Any
 from TTS.api import TTS
+import os
 
 # Internal imports
 from voices.voices import get_speaker_wav
@@ -16,7 +17,7 @@ def __init__(self) -> None:
         Initializes the VoiceCloner class
         """
 
-        self._tts = TTS('xtts')
+        self._tts = TTS('xtts-v2')
 
     def clone_voice(self, text: str) -> Any:
         """
@@ -55,6 +56,9 @@ def save_voice_locally(self, text: str, path: Path) -> None:
         - None
         """
 
+        if os.path.exists(str(path)):
+            os.remove(str(path))
+
         start_time = time.time() # Recoding how much tome it takes to clone the voice
         voice = self._tts.tts_to_file(
                 text=text,