Your voice is already responding to your words. You just didn't know it was happening.
Right now, as you write a script or narrate a document, Vois is reading the emotional temperature of what you're saying. When you write words like "amazing" or "incredible," the voice automatically speeds up—just slightly, just enough to feel energetic without sounding rushed. When you use words like "however" or "therefore," it hits the brakes and slows down just enough to let those moments land with weight.
This isn't magic. It's prosody—the subtle variation in speech that makes human voices sound natural instead of robotic. And Vois built it into the DNA of its engine without requiring you to touch a single SSML tag.
The Hidden Rhythm System
Most people don't think about words as having emotional weight. You just write what you want to say. But the words you choose carry emotional cargo, and your voice—both human and AI—picks up on it.
Vois's prosody system identifies specific keywords that signal emotional intensity or reflective thinking. Based on what it finds, it adjusts your playback speed by 4%—a subtle enough change that listeners don't consciously notice it, but dramatic enough that everything feels more natural.
Here's why this matters: when you speed up just slightly during exciting moments, the content feels more energetic. You're matching the emotion with the delivery. When you slow down during thoughtful, reflective language, you're giving those moments space to breathe. The listener feels the shift even if they can't articulate it.
Words That Speed You Up
These are the excitement words—language that signals positive emotion, celebration, discovery, or intensity. When Vois detects them, it bumps your delivery up 4% (1.04x speed).
The excitement trigger list:
- wow
- amazing
- incredible
- fantastic
- excellent
- wonderful
- brilliant
- awesome
- terrific
- fabulous
Use these in contexts where the energy matters. "This is an incredible breakthrough" lands harder and feels more genuine when the voice speeds up fractionally. A description of something wonderful gets the vocal energy it deserves.
Where does this help most? Product announcements. Discovery moments. Celebrations of achievement. Any moment where the human voice would naturally carry excitement, Vois matches that with a subtle speed increase.
Example: "The results were absolutely amazing—productivity increased by forty percent in just two weeks." The word "amazing" triggers the faster delivery, and suddenly that claim feels more convincing. The voice matches the magnitude of what you're claiming.
Words That Slow You Down
These are the thinking words—language that signals reflection, complexity, caution, or consequence. When Vois detects them, it backs off the gas and slows to 0.96x speed.
The reflective trigger list:
- however
- moreover
- therefore
- thus
- consequently
- nevertheless
- furthermore
- although
- meanwhile
- nonetheless
These words are conjunctions and connectors that do intellectual work. They're saying "hold on, there's more to this." "There's a contradiction here." "Let's think about what this means." They deserve space.
When you write "However, the implications were far more complex," that "however" signals the listener to lean in and pay attention. Slowing down by 4% gives that word and the clause it introduces genuine weight. Readers feel the pause even though you're still speaking.
Example: "The data showed improvement, however the long-term sustainability remained uncertain." That slowdown on "however" doesn't feel unnatural—it feels like the speaker is being thoughtful and careful with their words.
How to Use This Intentionally
Here's the thing: this system works automatically, but you can make it work better by being intentional about word choice.
Strategic Placement
Think about where you actually want shifts in energy. An audiobook chapter where a character makes an exciting discovery? Use excitement words at the moment of realization. "It was incredible—she finally understood." The speed bump makes that moment pop.
A documentary section where you're presenting important caveats or complications? Use reflective words to signal the shift. "Nevertheless, the situation proved more nuanced." The slowdown tells listeners this is important thinking, not casual information.
Building Natural Rhythm
The best scripts have rhythm. They speed up and slow down. They breathe. You can use emotional keywords to create this without ever thinking about technical pacing.
Instead of writing: "The results improved. There were limitations. We continued anyway."
Try: "The results were fantastic. However, limitations emerged. Yet we pushed forward."
The first version is flat. The second version has rhythm because you've given the engine natural prosody cues. The first sentence speeds up slightly (fantastic). The second slows down (however). The third accelerates again (we're continuing despite barriers). That's much closer to how a human would naturally tell that story.
When NOT to Use Emotion Words
Here's the trap: if you're trying to sound objective, clinical, or emotionally neutral, emotion words can work against you. If you're writing technical documentation or a news report where you want an even, measured tone, be sparing with excitement and reflection triggers.
Or use them deliberately. A documentary narrator breaking from clinical observation to express genuine amazement ("The discovery was absolutely incredible") creates a moment. It signals a shift in the narrator's perspective. But if you do this too often, you lose the power of the shift.
Combining with SSML for Full Control
Here's the advanced move: you don't have to choose between automatic emotion keywords and manual control. They work together.
You can use SSML to manually set exact speeds where you want them, and let the emotion keyword system handle everything else. Maybe you want a specific phrase to slow to 0.8x speed for dramatic effect—use SSML for that. But the day-to-day rhythm of your script? Let the emotion words do the work.
This was exciting! But <prosody rate="0.8">take a moment and really think about what this means</prosody>.
The excitement word ("exciting") triggers the automatic 4% speed increase. Your manual SSML on the second clause gives you exact control for dramatic effect. Best of both worlds.
Before and After: A Real Example
Let's look at a podcast script. First version with no attention to emotion keywords:
"The project launched successfully. We found unexpected complications. The team solved them quickly. Results exceeded expectations."
That's informative but flat. Even with natural speech, it feels like recitation.
Now the same script using emotion keywords intentionally:
"The project launched brilliantly. However, we encountered unexpected complications. The team solved them remarkably quickly. Results were absolutely fantastic—exceeding even our initial expectations."
Read the first version aloud. Then read the second. The second has energy. The slowdown on "however" makes the complications feel significant. The speedup on "brilliantly" and "fantastic" make the wins feel genuine. You can feel the rhythm even if you don't consciously notice the speed variations.
That's the whole point. You shouldn't have to think about prosody. You should just write naturally, use words that match the emotional content, and let Vois handle the rest.
The Bigger Picture
This is actually part of something larger: Vois's entire prosody system respects the emotional and intellectual structure of language. Emotion keywords are just one layer.
There's also punctuation handling (pauses), emphasis on key words, pacing adjustments based on sentence structure, and more. But the emotion keyword system is elegant because it requires nothing from you. You don't learn SSML. You don't adjust settings. You just write with some awareness that your word choices carry emotional information, and the voice responds naturally.
Most TTS systems treat all words equally. Vois understands that language isn't flat. Your words have meaning beyond their dictionary definitions. They carry emotion. They signal thinking. They build momentum or create space for reflection. And the voice should reflect that.
Getting Started
Next time you're writing a script or narrating content in Vois, try this: pick one moment where something exciting happens. Use words like "incredible" or "amazing." Then pick a moment where you want to signal caution or complexity. Use "however" or "nevertheless."
Generate the audio and listen. You'll hear the difference, even if it's subtle. That subtle difference is what makes the difference between robotic speech and human-sounding narration.
The emotion keyword system is already working in the background. Now you're just working with it instead of ignoring it.
Your scripts will sound better. Your audiobooks will feel more alive. Your podcasts will have rhythm. All because you're writing with awareness that words carry emotion, and voices respond to that emotion.
That's not artificial. That's just how language actually works—and Vois finally built an AI voice that understands it.