Skip to content

Repository files navigation

Description

This is the OpenVoiceOS TTS plugin for S.A.M. (Software Automatic Mouth).

SAM is a small Text-To-Speech (TTS) program written in C. It runs on most popular platforms. It is a C port of the SAM speech software for the Commodore C64, published in 1982 by Don't Ask Software (now SoftVoice, Inc.).

You can find a description in the original manual or in the manual of the equivalent Apple II program.

Install

pip install ovos-tts-plugin-SAM

Configuration

  "tts": {
    "module": "ovos-tts-plugin-SAM"
 }

The original manual lists these typical voice values:

DESCRIPTION          SPEED     PITCH     THROAT    MOUTH
Elf                   72        64        110       160
Little Robot          92        60        190       190
Stuffy Guy            82        72        110       105
Little Old Lady       82        32        145       145
Extra-Terrestrial    100        64        150       200
SAM                   72        64        128       128
  "tts": {
    "module": "ovos-tts-plugin-SAM",
    "ovos-tts-plugin-SAM": {
      "voice": "stuffy guy"
    }
 }

License

The software is a reverse-engineered version of a commercial program published more than 30 years ago. The current copyright holder is SoftVoice, Inc. (www.text2speech.com).

Attempts to contact the company failed. The website was last updated in 2009. The status of the original software is therefore best described as Abandonware.

As long as this is the case, the code cannot carry a specific open source software license. Use it at your own risk.

Docker

SAM runs behind ovos-tts-server (an ElevenLabs-compatible API) on port 9666. The image is fully offline: the retro sam binary is built into it, and it has no cloud dependency and no model download.

Pull the prebuilt image:

docker run -p 9666:9666 ghcr.io/openvoiceos/ovos-tts-plugin-sam:dev

...or build it locally:

docker build -t ovos-tts-plugin-sam .
docker run -p 9666:9666 ovos-tts-plugin-sam

...or use docker compose up.

Synthesize speech: http://localhost:9666/synthesize/hello

The default voice is SAM. Pick another one at build time (SAM, elf, little robot, stuffy guy, little old lady, extra-terrestrial):

docker build --build-arg SAM_VOICE="stuffy guy" -t ovos-tts-plugin-sam .

Related projects

About

S.A.M - Software Automatic Mouth

Topics

Resources

Contributing

Stars

0 stars

Watchers

3 watching

Forks

Releases

Sponsor this project

Packages

Used by

Contributors

Languages