Mms Tts Quz

🚀 Massively Multilingual Speech (MMS): Quechua, Cusco Text-to-Speech

This repository offers a text-to-speech (TTS) model checkpoint for the Quechua, Cusco (quz) language. It's part of Facebook's Massively Multilingual Speech project, which aims to offer speech technology across a wide range of languages. You can find more information about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview. All MMS-TTS checkpoints are available on the Hugging Face Hub: facebook/mms-tts. The MMS-TTS feature has been available in the 🤗 Transformers library since version 4.33.

✨ Features

• Multilingual Support: Part of a project targeting over 1000 languages.
• End - to - End Synthesis: Based on the VITS model for direct text - to - speech conversion.
• Stochastic Duration Prediction: Allows for speech synthesis with different rhythms from the same text.

📦 Installation

MMS-TTS is available in the 🤗 Transformers library starting from version 4.33. To use this checkpoint, first install the latest version of the library:

pip install --upgrade transformers accelerate

💻 Usage Examples

Basic Usage

from transformers import VitsModel, AutoTokenizer
import torch

model = VitsModel.from_pretrained("facebook/mms-tts-quz")
tokenizer = AutoTokenizer.from_pretrained("facebook/mms-tts-quz")

text = "some example text in the Quechua, Cusco language"
inputs = tokenizer(text, return_tensors="pt")

with torch.no_grad():
    output = model(**inputs).waveform

Saving the Output

import scipy

scipy.io.wavfile.write("techno.wav", rate=model.config.sampling_rate, data=output)

Displaying in Notebook

from IPython.display import Audio

Audio(output, rate=model.config.sampling_rate)

📚 Documentation

Model Details

VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) is an end-to-end speech synthesis model. It predicts a speech waveform based on an input text sequence. It's a conditional variational autoencoder (VAE) with a posterior encoder, decoder, and conditional prior.

A flow-based module, composed of a Transformer-based text encoder and multiple coupling layers, predicts a set of spectrogram-based acoustic features. The spectrogram is decoded using a stack of transposed convolutional layers, similar to the HiFi-GAN vocoder. Given the one-to-many nature of the TTS problem (where the same text can be spoken in multiple ways), the model includes a stochastic duration predictor. This allows the model to synthesize speech with different rhythms from the same input text.

The model is trained end-to-end using a combination of losses from variational lower bound and adversarial training. To enhance the model's expressiveness, normalizing flows are applied to the conditional prior distribution. During inference, text encodings are up-sampled according to the duration prediction module and then mapped to the waveform using a cascade of the flow module and HiFi-GAN decoder. Due to the stochastic nature of the duration predictor, the model is non-deterministic, so a fixed seed is needed to generate the same speech waveform.

For the MMS project, a separate VITS checkpoint is trained for each language.

📄 License

The model is licensed under CC-BY-NC 4.0.

📚 BibTex citation

This model was developed by Vineel Pratap et al. from Meta AI. If you use the model, consider citing the MMS paper:

@article{pratap2023mms,
    title={Scaling Speech Technology to 1,000+ Languages},
    author={Vineel Pratap and Andros Tjandra and Bowen Shi and Paden Tomasello and Arun Babu and Sayani Kundu and Ali Elkahky and Zhaoheng Ni and Apoorv Vyas and Maryam Fazel-Zarandi and Alexei Baevski and Yossi Adi and Xiaohui Zhang and Wei-Ning Hsu and Alexis Conneau and Michael Auli},
    journal={arXiv},
    year={2023}
}