SAM TTS Settings and Voice Parameter Lab

Hear what Speed, Pitch, Mouth and Throat change before moving the sliders. Load an exact example into the generator, compare it with the baseline, and save the recipe as a link or WAV.

SAM Settings Generator

Type one short line, preview it, then download the WAV.

93/900 charactersShort lines usually render clearer SAM-style audio.
Pronunciation help1 suggestion
Pronunciation helper
0:00 / 0:00

Download name: microsoft-sam-sam-tts.wav

Batch line rendering1/12 lines

Put each take on its own line, then download separate WAV files with the current voice and settings.

ReadySoftware Automatic Mouth (sam-js) | WAV export

A controlled parameter test

Use the sentence below for every take. Start with the classic baseline, then try the slower and lower variants. They change one parameter each, so differences are easier to attribute than with unrelated voice presets.

Classic baseline
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 example
Slow terminal
System 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 example
Lower robot
System 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 example

What the four controls actually do

Start at Speed 72, Pitch 64, Mouth 128 and Throat 128. These are raw SAM parameter values, not words per minute or musical notes. Change one control while keeping the text fixed.

Scroll the table sideways on smaller screens.

What the four controls actually do
ControlChangeWhat to check
SpeedIncrease for slower speech; decrease for faster speech.Compare 56, 72 and 96. More time can help separate words.
PitchIncrease for a lower voiced tone; decrease for a higher tone.Compare 48, 64 and 88. Mouth and Throat also affect the perceived character.
MouthChanges vocal resonance and vowel color.Compare 128 and 160 while keeping the other controls fixed.
ThroatChanges another part of the vocal resonance.Compare 128 and 160, then restore the baseline if words become unclear.

Fix the symptom before changing the character

For rushed words, increase Speed in small steps. For a wrong name, keep the voice fixed and edit the spelling. For harsh vowels, restore Mouth and Throat to 128. For a lower tone, increase Pitch rather than assuming the slider behaves like a conventional pitch control.

These recordings use the same engine as the generator, with no added processing. They demonstrate parameter differences; they are not ratings of voice quality or a promise that a recipe will suit every script.

Download and reuse the result

Generate the line, listen, then choose Download WAV. The current exporter produces mono, 22,050 Hz, 8-bit PCM WAV. It downloads generated speech, not a Windows installer or a voice package.

Keep a dry original before trimming or adding effects. If your editor or game requires 44.1 or 48 kHz, convert a copy there. Changing the format does not restore detail absent from the original synthesis. Copy Link preserves the text and four settings so you can reopen the take.

Which voice engine is used here?

This tool uses sam-js, a JavaScript port of Software Automatic Mouth. Microsoft Sam and Software Automatic Mouth are different speech systems. Labels such as Mike, Mary and RoboSoft identify this site's parameter presets; they do not load the original Microsoft or SAPI4 voice binaries.

Frequently asked questions

Why does a higher Speed value make SAM slower?

The interface passes the raw value to sam-js. It controls frame timing rather than words per minute, so a larger value lengthens speech.

Is Pitch measured in hertz?

No. It is a SAM engine parameter. Increasing it generally lowers the voiced tone; it is not a calibrated musical-note control.

Is there one best setting?

No. Keep a clear baseline and judge the final sentence in its intended mix. A dramatic preset can make instructions harder to understand.

Continue with a related task

Guide updated . Audio examples were generated with the displayed settings and no added effects.

Back to generator