← 返回命令列表

Linux command

tts 命令

文本

复制后可按需替换文件名、目录或参数。

常用示例

Synthesize speech

tts --text "[Hello world]"

Save to file

tts --text "[Hello]" --out_path [output.wav]

List models

tts --list_models

Use specific model

tts --text "[text]" --model_name [tts_models/en/ljspeech/tacotron2-DDC]

List speakers

tts --model_name [model] --list_speakers

说明

tts is the command-line interface for Coqui TTS, an open-source deep learning text-to-speech toolkit. It converts text input into natural-sounding speech audio using neural network models, supporting multiple languages and voice styles. The tool ships with a variety of pre-trained models that can be listed and selected by name. Multi-speaker models allow choosing between different voice characteristics via speaker indices. Audio output can be saved to WAV files or played directly. Models are downloaded automatically on first use and cached locally. While CPU inference is supported, a GPU is recommended for faster synthesis, especially with larger models.

参数

--text _TEXT_
Input text.
--out_path _FILE_
Output audio file.
--model_name _MODEL_
TTS model.
--list_models
Show models.
--speaker_idx _ID_
Speaker selection.

FAQ

What is the tts command used for?

tts is the command-line interface for Coqui TTS, an open-source deep learning text-to-speech toolkit. It converts text input into natural-sounding speech audio using neural network models, supporting multiple languages and voice styles. The tool ships with a variety of pre-trained models that can be listed and selected by name. Multi-speaker models allow choosing between different voice characteristics via speaker indices. Audio output can be saved to WAV files or played directly. Models are downloaded automatically on first use and cached locally. While CPU inference is supported, a GPU is recommended for faster synthesis, especially with larger models.

How do I run a basic tts example?

Run `tts --text "[Hello world]"` in a terminal, then adjust file names, paths, flags, or remote targets for your system.

What does --text _TEXT_ do in tts?

Input text.