app.py

import gradio as gr


from transformers import pipeline 

from datasets import load_dataset
import soundfile as sf
import torch
import os
os.environ['TRANSFORMERS_CACHE'] = '.cache'

print ("----- setting up pipeline -----")


synthesiser = pipeline("text-to-speech", "microsoft/speecht5_tts")

print ("----- setting up dataset -----")


embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
speaker_embedding = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
# You can replace this embedding with your own as well.

print ("----- synthetizing audio -----")

#speech = synthesiser("Hello, my dog is cooler than you!", forward_params={"speaker_embeddings": speaker_embedding})

#sf.write("speech.wav", speech["audio"], samplerate=speech["sampling_rate"])

def greet(name):
    return "Hello " + name + "!!"

def synthesise_audio(text, forward_params=None):
    speech = synthesiser(text, forward_params={"speaker_embeddings": speaker_embedding})
    sf.write("speech.wav", speech["audio"], samplerate=speech["sampling_rate"])
    return "speech.wav"

#demo = gr.Interface(fn=greet, inputs="text", outputs="text", description="----- TTS Testing -----")

demo = gr.Interface(fn=synthesise_audio,
                    inputs="text",
                    outputs="audio",
                    description="----- manuai Text To Speech generator -----",
                    allow_flagging = False)

demo.launch()