File size: 2,443 Bytes
eae52ca |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 |
import gradio as gr
import torch
import os
from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech, SpeechT5HifiGan
from datasets import load_dataset, Audio
import numpy as np
from speechbrain.inference import EncoderClassifier
# Load models and processor
processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
model = SpeechT5ForTextToSpeech.from_pretrained("tdnathmlenthusiast/speecht5_finetuned_techical_dataset_en")
vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
# Load speaker encoder
device = "cuda" if torch.cuda.is_available() else "cpu"
speaker_model = EncoderClassifier.from_hparams(
source="speechbrain/spkrec-xvect-voxceleb",
run_opts={"device": device},
savedir=os.path.join("/tmp", "speechbrain/spkrec-xvect-voxceleb")
)
def create_speaker_embedding(waveform):
with torch.no_grad():
speaker_embeddings = speaker_model.encode_batch(torch.tensor(waveform))
speaker_embeddings = torch.nn.functional.normalize(speaker_embeddings, dim=2)
speaker_embeddings = speaker_embeddings.squeeze().cpu().numpy()
return speaker_embeddings
# Load a sample from the dataset for speaker embedding
try:
dataset = load_dataset("Yassmen/TTS_English_Technical_data", split="train", trust_remote_code=True)
dataset = dataset.cast_column("audio", Audio(sampling_rate=16000))
sample = dataset[50]
speaker_embedding = create_speaker_embedding(sample['audio']['array'])
except Exception as e:
print(f"Error loading dataset: {e}")
# Use a random speaker embedding as fallback
speaker_embedding = torch.randn(1, 512)
def text_to_speech(text):
# Clean up text
replacements = [
("0", "zero"),
("1", "one"),
("2", "two"),
("3", "three"),
("4", "four"),
("5", "five"),
("6", "six"),
("7", "seven"),
("8", "eight"),
("9", "nine"),
("_", " ")
]
for src, dst in replacements:
text = text.replace(src, dst)
inputs = processor(text=text, return_tensors="pt")
speech = model.generate_speech(inputs["input_ids"], speaker_embedding, vocoder=vocoder)
return (16000, speech.numpy())
iface = gr.Interface(
fn=text_to_speech,
inputs="text",
outputs="audio",
title="Technical English Text-to-Speech",
description="Enter english text to convert to speech"
)
iface.launch() |