Anexdeus/Roxy_Migurdia_coqui_XTTS

ⓍTTS trained on Roxy's voice

How to use

Init from XttsConfig and load checkpoint:

import torchaudio
from TTS.tts.configs.xtts_config import XttsConfig
from TTS.tts.models.xtts import Xtts

SPEAKER_REFERENCE = "reference.wav"

config = XttsConfig()
config.load_json("config.json")
model = Xtts.init_from_config(config)
model.load_checkpoint(config, checkpoint_path="best_model_180.pth", vocab_path="vocab.json", use_deepspeed=False)
model.cuda()

gpt_cond_latent, speaker_embedding = model.get_conditioning_latents(audio_path=[SPEAKER_REFERENCE]) # Use sample of Roxy's voice here

out = model.inference(
    "It took me quite a long time to develop a voice and now that I have it I am not going to be silent.",
    "en",
    gpt_cond_latent,
    speaker_embedding,
    temperature=0.7, # Add custom parameters here
)
torchaudio.save("output.wav", torch.tensor(out["wav"]).unsqueeze(0), 24000)