Model Overview

This collection contains two models:

1. Single-speaker FastPitch (around 50M parameters) trained on SF Chinese/English Bilingual Speech dataset [1].
2. HiFiGAN trained on Mel-spectrograms predicted by the FastPitch in (1).

Model Architecture

FastPitch [2] is a non-autoregressive model for mel-spectrogram generation based on FastSpeech [3], conditioned on fundamental frequency contours. It uses an external Tacotron 2 [4] model trained on LJSpeech-1.1 to extract training alignments, and estimate durations of input symbols. NeMo implemetation leverages a novel alignment framework [5] to simplify the alignment learning in TTS models. For more informaiton on training a FastPitch model, please refer to NeMo tutorial FastPitch_MixerTTS_Training.

HiFiGAN [6] is a generative adversarial network (GAN) model that generates audios from mel-spectrograms. The generator uses transposed convolutions to upsample mel-spectrograms to audios. For more details about HiFiGAN, please refer to its original paper. NeMo re-implementation of HiFiGAN can be found here.

Training

You can follow the NeMo Chinese TTS training tutorial for details: https://github.com/NVIDIA/NeMo/blob/r1.14.0/tutorials/tts/FastPitch_ChineseTTS_Training.ipynb

Datasets

• FastPitch: the model is trained from scratch on SF Chinese/English Bilingual Speech dataset [1] sampled at 22050 Hz. The dataset contains about 2,740 bilingual audio samples of a single female speaker and their corresponding text transcripts, each of them is an audio of around 5-6 seconds and have a total length of approximately 4.5 hours.
• HiFiGAN: the model is finetuned from the tts_en_lj_hifigan_ft_mixerttsx.nemo in TTS En LJ HiFi-GAN by the mel-spectrograms generated from the FastPitch model above.

Performance

No performance information available at this time.

How to Use this Model

The model is available for use in the NeMo toolkit [7], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.

NOTE: For best results you should use the vocoder (HiFiGAN) checkpoint in this model card along with the mel spectrogram generator (FastPitch) checkpoint.

Automatically load the model from NGC

# Load spectrogram generator
from nemo.collections.tts.models import FastPitchModel
spec_generator = FastPitchModel.from_pretrained(model_name="tts_zh_fastpitch_sfspeech")

# Load Vocoder
from nemo.collections.tts.models import HifiGanModel
model = HifiGanModel.from_pretrained(model_name="tts_zh_hifigan_sfspeech")

# Generate audio
import soundfile as sf
import torch
with torch.no_grad():
    parsed = spec_generator.parse("这些新一代的CPU不只效能惊人。")
    spectrogram = spec_generator.generate_spectrogram(tokens=parsed)
    audio = model.convert_spectrogram_to_audio(spec=spectrogram)
    if isinstance(audio, torch.Tensor):
        audio = audio.to('cpu').numpy()

# Save the audio to disk in a file called speech.wav
sf.write("speech.wav", audio.T, 22050, format='WAV')

Input

This model accepts batches of texts.

Output

This model generates mel spectrograms.

Limitations

Since this model was trained on publically available speech datasets, the performance of this model might degrade for speech which includes technical terms, or vernacular that the model has not been trained on. The model might also perform worse for accented speech.

Versions

• 1.15 (latest): Disambiguate polyphones with augmented dict and Jieba segmenter.
• 1.14: Initial models.

References

• [1] SF Chinese/English Bilingual Speech Dataset: https://catalog.ngc.nvidia.com/orgs/nvidia/resources/sf_bilingual_speech_zh_en
• [2] FastPitch: Parallel Text-to-speech with Pitch Prediction: https://arxiv.org/abs/2006.06873
• [3] Fastspeech: Fast, robust and controllable text to speech: https://arxiv.org/abs/1905.09263
• [4] Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions https://arxiv.org/abs/1712.05884
• [5] One TTS Alignment To Rule Them All: https://arxiv.org/abs/2108.10447
• [6] HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis: https://arxiv.org/abs/2010.05646
• [7] NVIDIA NeMo Toolkit: https://github.com/NVIDIA/NeMo

NVIDIA License

1. Definitions

"Licensor" means any person or entity that distributes its Work.

"Work" means (a) the original work of authorship made available under this license, which may include software, documentation, or other files, and (b) any additions to or derivative works thereof that are made available under this license.

The terms "reproduce," "reproduction," "derivative works," and "distribution" have the meaning as provided under U.S. copyright law; provided, however, that for the purposes of this license, derivative works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work.

Works are "made available" under this license by including in or with the Work either (a) a copyright notice referencing the applicability of this license to the Work, or (b) a copy of this license.

2. License Grant

2.1 Copyright Grant. Subject to the terms and conditions of this license, each Licensor grants to you a perpetual, worldwide, non-exclusive, royalty-free, copyright license to use, reproduce, prepare derivative works of, publicly display, publicly perform, sublicense and distribute its Work and any resulting derivative works in any form.

3. Limitations

3.1. Redistribution. You may reproduce or distribute the Work only if (a) you do so under this license, (b) you include a complete copy of this license with your distribution, and (c) you retain without modification any copyright, patent, trademark, or attribution notices that are present in the Work.

3.2 Derivative Works. You may specify that additional or different terms apply to the use, reproduction, and distribution of your derivative works of the Work ("Your Terms") only if (a) Your Terms provide that the use limitation in Section 3.3 applies to your derivative works, and (b) you identify the specific derivative works that are subject to Your Terms. Notwithstanding Your Terms, this license (including the redistribution requirements in Section 3.1) will continue to apply to the Work itself.

3.3 Use Limitation. The Work and any derivative works thereof only may be used or intended for use non-commercially. Notwithstanding the foregoing, NVIDIA Corporation and its affiliates may use the Work and any derivative works commercially. As used herein, "non-commercially" means for research or evaluation purposes only.

3.4 Patent Claims. If you bring or threaten to bring a patent claim against any Licensor (including any claim, cross-claim or counterclaim in a lawsuit) to enforce any patents that you allege are infringed by any Work, then your rights under this license from such Licensor (including the grant in Section 2.1) will terminate immediately.

3.5 Trademarks. This license does not grant any rights to use any Licensor's or its affiliates' names, logos, or trademarks, except as necessary to reproduce the notices described in this license.

3.6 Termination. If you violate any term of this license, then your rights under this license (including the grant in Section 2.1) will terminate immediately.

4. Disclaimer of Warranty.

THE WORK IS PROVIDED "AS IS" WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF

MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER THIS LICENSE.

5. Limitation of Liability.

EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK (INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION, LOST PROFITS OR DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER DAMAGES OR LOSSES), EVEN IF THE LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.