Hobbyist Runs 1982 Speech Synthesis Software On A 1977 Atari 2600

Hobbyist Runs 1982 Speech Synthesis Software On A 1977 Atari 2600

Reviving nostalgia through sounds might be a primary source of joy, which could just boost your mood or make you feel sentimental immediately. A blogger obsessed about merging vintage game consoles and computers with more recent technologies has brought up his Atari 2600 computer to the show. In this case, he merged an old speech software with an even older computer which surprisingly worked.

A 1982 software on a 1977 computer

The Software Automatic Mouth (SAM) is a talking and singing software introduced in 1982. The voice synthesis program was particularly designed for Atari, Apple, and Commodore computers and had a mechanism that converts text to phonemes and phonemes to speech. You could just think of it as an older version of Siri, different sounds are added onto each other which then form into sentences.


The blogger, who is seemingly in love with Atari computers, had this thought of running SAM on Atari 2600, a technology from the year 1977. SAM worked on Atari 800, which was released in 1979 and manufactured until 1992, it had a relatively huge RAM of 48K. However, the capacity of Atari 2600 was only 128 bytes.

How did it work?

After 38 years of the software's release, the blogger fit SAM into Atari 2600 to run. He took a compressed stream of phoneme, allophone, and timing data which is run by a code and moved it to a web-based tool.

"The SAM2600 Authoring Tool produces a compressed format designed to be compact but easy to interpret at runtime. Individual phonemes and allophones are encoded as two or three bytes. Pauses are encoded as 1 byte. This format encodes speech at around 25-30 bytes per second, roughly the speed of a 300 baud modem." he explained the complex process on his blog.

The blogger recorded its project on a video and posted it on YouTube with a caption "Make your 2600 talk! Sing! Say rude words!" However, make sure you turn on the subtitles, as it is a bit hard to understand and sounds more like a primitive robot responding to one-command tasks assigned by its user.

Watch the video: . 2600 (May 2021).