Model Overview

FastPitchHifiGanE2E is an end-to-end, non-autoregressive model that generates audio from text. It combines FastPitch and HiFiGan into one model and is traned jointly in an end-to-end manner.

Model Architecture

The FastPitch portion consists of the same transformer-based encoder, pitch predictor, and duration predictor as the original FastPitch model. The HiFiGan portion takes the discriminator from HiFiGan and uses it to generate audio from the output of the FastPitch portion. No spectrograms are used in the training of the model. All losses are taken from HiFiGan plus additional losses for the pitch and duration predictors.

Training

This model is trained on LJSpeech sampled at 22050Hz, and has been tested on generating female English voices with an American accent.

Performance

No performance information available at this time.

How to Use this Model

This model can be automatically loaded from NGC.

import soundfile as sf
from nemo.collections.tts.models import FastPitchHifiGanE2EModel

# Load the model from NGC
model = FastPitchHifiGanE2EModel.from_pretrained(model_name="tts_en_e2e_fastpitchhifigan")

# Run inference
tokens = model.parse("Hey, I can speak!")
audio = model.convert_text_to_waveform(tokens=tokens)

# Save the audio to disk in a file called speech.wav
sf.write("speech.wav", audio.to('cpu').numpy(), 22050)

Input

This model accepts batches of text.

Output

This model generates audio.

Limitations

This model outputs audio at 22050Hz.

Versions

1.0.0 (current): The original version released with NeMo 1.0.0.

Licence

License to use this model is covered by the NGC TERMS OF USE unless another License/Terms Of Use/EULA is clearly specified. By downloading the public and release version of the model, you accept the terms and conditions of the NGC TERMS OF USE.