Skip to contents

Convert Text to Speech

Source: R/multimodal.R
hf_text_to_speech.Rd

Generate speech audio from text and write it to disk using a text-to-speech model.

Usage

hf_text_to_speech(
  text,
  output = NULL,
  model = hf_default_model("text_to_speech"),
  token = NULL,
  endpoint_url = NULL,
  overwrite = FALSE,
  ...
)

Arguments

text

Character vector of text to synthesize.

output

Character path(s) or NULL. When NULL, files are written to temporary paths with an extension inferred from the response content type.

model

Character string. Model ID from Hugging Face Hub. Default: "hexgrad/Kokoro-82M".

token

Character string or NULL. API token for authentication.

endpoint_url

Character string or NULL. A custom Inference Endpoint URL.

overwrite

Logical. If TRUE, overwrite existing output files.

...

Additional generation parameters passed to the model.

Value

A tibble with columns: text, path, content_type, audio.

Provider availability

huggingfaceR's task functions speak the Hugging Face task API contract, which only the first-party `hf-inference` provider implements. At the time of writing `hf-inference` serves no text-to-speech model, so serverless text-to-speech is not reachable through this function: calls raise an error naming the third-party providers that do serve the model. To synthesise speech today, deploy the model as a dedicated Inference Endpoint and pass `endpoint_url`.

See also

https://huggingface.co/tasks/text-to-speech

Examples

if (FALSE) { # \dontrun{
hf_text_to_speech("Hello from R.")
} # }