chatterbox

Text-to-Speech Using Chatterbox TTS Engine

Last updated: 2026-01-30

chatterbox is an R package that is an R port of resemble AI’s chatterbox library. It is written entirely in R using torch and has no Python dependencies.

Installation

You can install the development version of chatterbox from GitHub with:

1remotes::install_github("cornball-ai/chatterbox")

Usage

 1# Set timeout to 10 minutes to allow model download
 2options(timeout = 600)
 3
 4library(chatterbox)
 5
 6# Load model
 7model <- chatterbox("cuda")
 8model <- load_chatterbox(model)
 9
10# Generate speech
11jfk <- system.file("audio", "jfk.mp3", package = "chatterbox")
12result <- tts(model, "Hello, this is a test!", jfk)
13write_audio(result$audio, result$sample_rate, "output.wav")
14
15# Or one-liner:
16quick_tts("Hello world!", "ref.wav", "out.wav")

Reference

See Function Reference for complete API documentation.

Functions