mms-tts

Massively Multilingual Speech (MMS) : Text-to-Speech Models

This repository contains a collection of text-to-speech (TTS) models, offering support for over 1000 languages. These models are part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages.

Table Of Contents

• Usage
• Supported Languages
• Model details
• Additional links

Usage

For detailed instructions on utilizing the models, please refer to the fairseq docs. Additionally, you can explore the models on MMS Space available on HuggingFace.

models folder consists of only generator, which is enough to run TTS inference. Full model checkpoint which also includes the discriminator and the optimizer states can be found in full_models folder. The models can be downloaded locally using hf_hub_download API.

See this section for instructions on using the models for inference.

Supported Languages

This model supports 1107 languages. Unclick the following to toogle all supported languages of this checkpoint in ISO 639-3 code. You can find more details about the languages and their ISO 639-3 codes in the MMS Language Coverage Overview.

Model details

• Developed by: Vineel Pratap et al.

• Model type: Text-to-speech model

• Language(s): 1107 languages, see supported languages

• License: CC-BY-NC 4.0 license

• Cite as:

  @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}
  }
  

Additional Links

• Blog post
• Transformers documentation.
• Paper
• GitHub Repository
• Other MMS checkpoints
• MMS base checkpoints:
  • facebook/mms-1b
  • facebook/mms-300m
• Official Space