done: json file works

voice_module.py

@@ -17,16 +17,15 @@
 
 
 def audio_to_text(filename):
-    response = get_att_whisper(filename, 'txt')
+    response = get_att_whisper(filename)
 
     if response.status_code == 200:
-        transcription = filename.split('.')[0] + '.json'
-        result_json = json.dumps(transcription)
-        recognized_data = json.loads(result_json)
 
-        input_sentence = RecognizedSentence(recognized_data)
+        input_sentence = RecognizedSentence(response.json())
         return input_sentence
 
+    return None
+
 
 def download_voice(update: Update):
     downloaded_file = update.message.voice.get_file()

whisper_module.py

@@ -9,7 +9,7 @@ class WhisperSettings:
     asr_model = ASR_MODEL
 
 
-def get_att_whisper(filename, output: str = 'json'):
+def get_att_whisper(filename):
     """
      filename: str
         audio file name
@@ -19,8 +19,11 @@ def get_att_whisper(filename, output: str = 'json'):
             "txt", "vtt", "srt", "tsv", "json"
     """
 
-    file = {'audio_file': (filename, open(filename, 'rb'))}
-    data = {'task': 'transcribe', 'output': output, 'word_timestamps': 'true'}
-    response = requests.post(WhisperSettings.link + '/asr', data=data, files=file)
+    with open(filename, 'rb') as f_byte:
+        file = {'audio_file': (filename, f_byte)}
+
+    response = requests.post(WhisperSettings.link +
+                             '/asr?task=transcribe&encode=false&output=json&word_timestamps=true',
+                             files=file)
 
     return response