I kept the default to `alloy`
@ -14,7 +14,7 @@ class Tts:
def tts(self, text):
response = client.audio.speech.create(
model="tts-1",
voice="alloy",
voice=os.getenv('OPENAI_VOICE_NAME', 'alloy'),
input=text,
response_format="opus"
)