FastSpeech2-en-200_Speaker-cv4: An Open-Source English Text-to-Speech Model

Fastspeech2 En 200 Speaker Cv4

Developed by facebook

An English text-to-speech model based on the FastSpeech 2 architecture, supporting 200 different voices, trained on the Common Voice v4 dataset.

Speech Synthesis English#Multi-speaker TTS #English speech synthesis #FastSpeech2 architecture

Downloads 37

Release Time : 3/2/2022

Model Overview

This is a multi-speaker text-to-speech model capable of converting English text into natural speech, supporting 200 different male and female voices.

Model Features

Multi-speaker support

The model supports 200 different male and female voices, allowing random speaker selection during use.

High-quality speech synthesis

Based on the FastSpeech 2 architecture, it can generate natural and fluent speech output.

Large-scale dataset training

Trained on the Common Voice v4 dataset, ensuring the model's generalization capability.

Model Capabilities

English text-to-speech

Multi-speaker speech synthesis

Use Cases

Speech synthesis applications

Voice assistants

Provides natural multi-voice speech output for voice assistant systems.

Generates natural and fluent speech responses

Audiobooks

Automatically converts text content into audiobooks with multiple voices.

Supports reading in 200 different voices

🚀 fastspeech2-en-200_speaker-cv4

A text-to-speech model based on FastSpeech 2 from fairseq S^2, capable of generating English speech with 200 different voices.

🚀 Quick Start

This is a FastSpeech 2 text-to-speech model from fairseq S^2 (paper/code). It has the following characteristics:

Language: English
Voices: 200 male/female voices (random speaker when using the widget)
Training Data: Trained on Common Voice v4

📦 Installation

No specific installation steps are provided in the original document.

💻 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/fastspeech2-en-200_speaker-cv4",
    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 = "Hello, this is a test run."

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

ipd.Audio(wav, rate=rate)

Advanced Usage

📚 Documentation

Property	Details
Model Type	FastSpeech 2 text-to-speech model
Training Data	Common Voice v4

📄 License

No license information is provided in the original document.

📚 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",
}

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