pythonpygametext-to-speechpydubgtts

Pygame mixer.music.load() does not work with gTTs

I'm trying to play a gTTs voice with pygame.mixer.music.load() only. I don't want to save the voice into a file, so I saved it into a BytesIO stream. gTTs returns .mp3 audio which I know has limited support by pygame, so I tried to convert the .mp3 audio to .wav using the pydub module, but I couldn't find a way to do so without saving it to a file. How can I fix this issue in any way possible?

from pygame import mixer
from gtts import gTTS

def play(buffer):
   buffer.seek(0)
   mixer.music.load(buffer) #Load the mp3
   print("Sound loaded. Time to play!")
   mixer.music.play() #Play it

def generate_voice(text, accent):
   mp3_fp = BytesIO()
   tts = gTTS(text)
   tts.write_to_fp(mp3_fp)
   return mp3_fp

text = "Hi there"
buffer = generate_voice(text, accent)
play(buffer)

The error returned by pygame.mixer.music.load(): pygame.error: ModPlug_Load failed

Solution

  • I fixed this issue by using pydub to convert the audio into a wav format:

    def play(buffer):
   mixer.init() 
   mixer.music.load(buffer) #Load the mp3
   print("Sound loaded. Time to play!")
   mixer.music.play() #Play it

def generate_voice(text, lang):
   fp = BytesIO()
   wav_fp = BytesIO()
   tts = gTTS(text=text, lang=lang)
   tts.write_to_fp(fp)
   fp.seek(0)
   sound = AudioSegment.from_file(fp)
   wav_fp = sound.export(fp, format = "wav")
   return wav_fp