tts_transformer-es-css10 open-source text-to-speech model - Free deployment supports Spanish speech synthesis

Tts Transformer Es Css10

Developed by facebook

Fairseq S^2-based Transformer text-to-speech model supporting Spanish, trained on CSS10 dataset

Speech Synthesis Spanish#Spanish TTS #Single male speaker #Transformer architecture

Downloads 50

Release Time : 3/2/2022

Model Overview

This is a Transformer architecture-based text-to-speech model specifically designed to convert Spanish text into natural speech. The model is trained on the CSS10 dataset and generates a single male speaker voice.

Model Features

Transformer-based architecture

Utilizes advanced Transformer architecture to deliver high-quality speech synthesis

Spanish language support

Speech synthesis model specifically optimized for Spanish

Single speaker model

Generates consistent male voice output

fairseq integration

Part of the fairseq S^2 toolkit, easy to integrate and use

Model Capabilities

Spanish text-to-speech

High-quality speech synthesis

Real-time speech generation

Use Cases

Voice applications

Voice assistants

Provides natural voice output for Spanish voice assistants

Generates natural and fluent Spanish speech

Audiobooks

Converts Spanish text into audiobooks

Produces comfortable voice suitable for extended listening

Education

Language learning

Provides pronunciation examples for Spanish learners

Generates standard Spanish pronunciation

🚀 tts_transformer-es-css10

A Transformer-based text-to-speech model from fairseq S^2, capable of converting Spanish text into a single-speaker male voice, trained on the CSS10 dataset.

🚀 Quick Start

This is a Transformer text-to-speech model from fairseq S^2 (paper/code):

Supports Spanish.
Generates a single-speaker male voice.
Trained on CSS10.

📦 Installation

Dependencies

pip install fairseq sentencepiece

💻 Usage Examples

Basic Usage

from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd


models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
    "facebook/tts_transformer-es-css10",
    arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator([model], cfg)

text = "Hola, esta es una prueba."

sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)

ipd.Audio(wav, rate=rate)

📄 License

Citation

@inproceedings{wang-etal-2021-fairseq,
    title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
    author = "Wang, Changhan  and
      Hsu, Wei-Ning  and
      Adi, Yossi  and
      Polyak, Adam  and
      Lee, Ann  and
      Chen, Peng-Jen  and
      Gu, Jiatao  and
      Pino, Juan",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.emnlp-demo.17",
    doi = "10.18653/v1/2021.emnlp-demo.17",
    pages = "143--152",
}

📋 Information Table

Property	Details
Library Name	fairseq
Task	text-to-speech
Tags	fairseq, audio, text-to-speech
Language	Spanish
Datasets	CSS10

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご