ETS2 custom GPS voice: make your own navigation voice mod
Euro Truck Simulator 2 and American Truck Simulator let modders replace the voice of the navigation. Players make navigation voices in other languages, in a friend’s voice or in a game or movie character’s voice. The hard part is recording dozens of short commands in one consistent voice.
This guide lists the commands the game expects according to the SCS modding wiki, shows how to voice all of them at once with the free SuperClonica engine, and points to what you need to package them for the current game versions.
The lines the navigation needs
The SCS modding wiki lists the sound names the navigation uses: turn_left, turn_right, keep_left, keep_right, exit_left, exit_right and go_straight; roundabout_1 to roundabout_6 and exit_now; prepare_turn_left, prepare_turn_right, prepare_exit_left and prepare_exit_right; compound_ and and_then_ versions for two commands in a row; and start, finish, recomputing, u_turn and speed_warning. Write one natural sentence for each, such as “In two hundred metres, turn left” for prepare_turn_left.
Pick a voice and voice every line
Hear voices free on their pages with one command first, up to 100 characters, no account needed; character voices make fun packs, narrators make realistic ones. Then put the names and sentences in a table with two columns, name and line, install the free SuperClonica engine with python -m pip install superclonica-cpu and, in Python, after import csv, from superclonica import Superclonica and tts = Superclonica(), then for row in csv.DictReader(open("navigation.csv", encoding="utf-8")): audio, rate = tts.synthesize(row["line"], voice="navigator.json", lang="en", seed=4); tts.save(row["name"] + ".wav", audio). A fixed seed keeps the voice the same across all commands.
Packaging for the current game
The wiki page describes a pack with a configuration file in /base/sound/navigation, a pack_name shown in the game options and a pack_dir with the sounds. According to the SCS forum, since game version 1.37 the sounds go into an FMOD sound bank built with the FMOD Studio version recommended on the SCS wiki, and event names must match the wiki list exactly. Players also warn to pack the mod archive without compression.
Tips for a clean navigation voice
Keep commands short, with distances written as words, and keep the same calm tone in all of them. Use --normalize so every command has the same loudness, and trim silence at the start so the voice speaks right when the game triggers it. The engine speaks 31 languages, so you can make a pack for your own language; Chinese is not supported yet.
Fair use of voices
Character voices in the catalog are fan-made; share packs made with them as free fan mods, not as official content. For a voice of a real friend, get their consent, or train your own voice in the studio from a clean 5 to 10 second recording; training on your own NVIDIA graphics card is free.