RHVoice

RHVoice is a multilingual speech synthesizer. It uses statistical parametric synthesis and relies on existing open-source speech technologies (mainly HTS and related software).

Installation

Install the rhvoice package. Also install necessary #Languages and #Voices.

Languages

Initially, RHVoice could speak only Russian. But now it supports many other languages.

Available languages list you can get by command:

$ pacman -Ss rhvoice-lang

Voices

Voices are built from recordings of natural speech. They have small footprints, because only statistical models are stored on users' computers. And though the voices lack the naturalness of the synthesizers which generate speech by combining segments of the recordings themselves, they are still very intelligible and resemble the speakers who recorded the source material.

Available voices list you can get by command:

$ pacman -Ss rhvoice-voice

Configuration

Configuration file located at: /etc/RHVoice/RHVoice.conf.

File format and available settings are detail described in official documentation.

/etc/speech-dispatcher/speechd.conf

# -----OUTPUT MODULES CONFIGURATION-----
...
DefaultModule rhvoice
...

Usage

If configuration you made is correct, the following commands will allow you to start the synthesis (for voice bdl):

$ echo "test" | RHVoice-test -p "bdl"
$ spd-say -o rhvoice -y bdl "test"

Dictionaries

User dictionaries must be created in the /etc/RHVoice/dicts/ directory, for example, for the English language: /etc/RHVoice/dicts/English/.

See also

• RHVoice home page.
• Project RHVoice in GitHub.
• Official documentation.
• RHVoice Lab - Lab for creating new voices