Best SAM TTS Settings for Five Robot Voice Styles
Hear five SAM robot recipes with exact parameters. Compare pacing, pitch and resonance, then load the same examples into the voice generator.
In this guide
Five recipes with the same test sentence
Use these five starting points to compare a classic baseline, slower terminal, lower robot, higher robot and resonance change. Every recording says System ready. Proceed to platform three. Load an example to reproduce its settings in the generator.
The samples are generated by the site's sam-js engine with no added effects. They are controlled examples, not measurements of perceived quality. Choose a recipe using your actual line and playback conditions.
System ready. Proceed to platform three.Speed 72 · Pitch 64 · Mouth 128 · Throat 128
Reference take. Compare every change against this same sentence.
Load this exampleSystem ready. Proceed to platform three.Speed 96 · Pitch 64 · Mouth 128 · Throat 128
Only Speed changes. A larger value lengthens the spoken line.
Load this exampleSystem ready. Proceed to platform three.Speed 72 · Pitch 88 · Mouth 128 · Throat 128
Only Pitch changes. A larger value lowers the voiced tone in this engine.
Load this exampleSystem ready. Proceed to platform three.Speed 72 · Pitch 48 · Mouth 128 · Throat 128
Only Pitch changes. A smaller value raises the voiced tone.
Load this exampleSystem ready. Proceed to platform three.Speed 72 · Pitch 64 · Mouth 190 · Throat 190
Pitch and timing stay at the baseline; Mouth and Throat change the vowel character.
Load this exampleRead the parameter direction correctly
Speed is an engine timing value: a larger number makes speech slower. Pitch is also a raw engine parameter: a larger number generally lowers the voiced tone. Neither control is calibrated in words per minute, hertz or musical notes.
Mouth and Throat change resonance. They can alter the perceived size and color of a robot without doing the same thing as the Pitch control. A preset that changes all four values can therefore sound quite different from a pitch-only experiment.
Scroll the table sideways on smaller screens.
| Recipe | Speed / Pitch / Mouth / Throat | Change from baseline |
|---|---|---|
| Classic | 72 / 64 / 128 / 128 | Reference |
| Slow terminal | 96 / 64 / 128 / 128 | Speed only |
| Lower robot | 72 / 88 / 128 / 128 | Pitch only |
| Higher robot | 72 / 48 / 128 / 128 | Pitch only |
| Resonance | 72 / 64 / 190 / 190 | Mouth and Throat |
Choose by what the line needs to communicate
For a terminal instruction, start with the baseline and slower take. Check whether platform three is easy to catch on the first play. For a character cue, compare the two pitch variants before changing the resonance. For a machine panel, try the resonance example and check that the consonants remain readable.
The five recipes are deliberately modest. Do not assume the most exaggerated voice is the most useful one. A long status message and a one-word alert may need separate timing even when they share the same character.
Move from the sample to your script
Load the chosen example, replace only the text and generate again. Keep a copy of the sample recipe before adjusting the new line. If one word fails, fix that spelling first; changing the voice to solve a single name can make every other word worse.
Save the accepted result with its four values or a copied settings link. For a game with many robot lines, use the same Mouth, Throat and Pitch as a character baseline and make only small timing changes when necessary.
Add radio or room effects after synthesis
The parameters change the synthesizer, not an effects rack. They do not add a telephone band-pass filter, static, room reverb or an echo. Export a dry WAV before adding those layers in an editor.
Compare the processed clip against the dry source at similar playback levels. If an effect hides a warning word, reduce the effect or simplify the sentence before changing the character recipe.