gemini-2.5-pro-tts-voice-profiles-dramabox

# Gemini 2.5 Pro TTS Voice Profiles — with DramaBox prompts This is [laion/gemini-2.5-pro-tts-voice-profiles](https://huggingface.co/datasets/laion/gemini-2.5-pro-tts-voice-profiles) **augmented with two DramaBox performance prompts per sample**, generated by `google/gemma-3-12b-it` from the dataset's existing annotations (BUD-E Whisper caption, Empathic-Insight scores, Gemini word-level transcriptions/captions, and vocal-burst descriptions) — no re-running of ASR or whisper experts. Everything from the original dataset is preserved; each sample's `.json` gains: | Field | Description | |-------|-------------| | `dramabox_prompt_full` | DramaBox performance prompt **with** speaker demographics (age, gender, timbre) + emotional stage directions + dialogue. Vocal bursts woven into the stage directions. | | `dramabox_prompt_focused` | DramaBox prompt **without** demographics — pure voice-quality + emotional delivery + dialogue (for reference-audio / unbiased TTS). | | `dramabox_model` | The generation model id. | | `dramabox_meta.full_starter` / `focused_starter` | The randomized opening word/phrase used (opening-line diversity so prompts don't all begin identically). | Prompt format: a speaker/voice description paragraph ending in `"delivering this high-quality studio voice recording with no background noise."`, then alternating `(stage directions)` and `"dialogue"`. ## Example prompts ### FULL (with speaker demographics) **Charon** · en (speaking with a US accent) · emotion: superiority/mournfulness · bursts: breath_catch, approving_mmm ``` Listen — A man, appearing to be in his early thirties, with a low, resonant voice and a slightly bitter timbre, delivering this high-quality studio voice recording with no background noise. (He sits at a table, nursing a bowl of noodles, a faint sadness clouding his eyes. A brief breath_catch as a memory surfaces.) “Nobody makes noodles like my grandmother used to. Nobody.” (His voice drops slightly, laced with a nostalgic longing. He pauses, his gaze fixed on the noodles.) “They just… understood what it meant to nurture.” (A small, almost involuntary approving_mmm escapes his lips as he tastes the broth.) “This broth... it's good, really.” (He looks up, a flicker of resignation crossing his face.) “But it's just not *hers*, is it?” ``` **Kore** · en (speaking with a US accent) · emotion: sullenness/jealousy & envy · bursts: teeth_click, wistful_sigh ``` A voice — A young adult female, approximately 27 years old, with a low-pitched, drawn-out delivery and a slightly detached timbre, delivering this high-quality studio voice recording with no background noise. (She stands in a dimly lit room, shoulders slumped slightly, a subtle frown creasing her brow. A flicker of annoyance crosses her face.) "Zoning out again, I see." (A brief pause, a tightening of her lips.) “I just… it's fine, honestly.” (She looks down, avoiding eye contact, a slight tremor in her voice as she continues.) “He's just, you know, *there* all the time.” (A short clack from lightly tapping teeth – teeth_click – punctuates the sentence, a nervous habit betraying her composure.) “It’s not like she even *notices* me anymore.” (Her voice cracks slightly, a wistful sigh – wistful_sigh – escapes her as she looks off into the distance, lost in thought.) “I guess they’re happy.” (She shakes her head almost imperceptibly, a flicker of jealousy in her eyes.) “Whatever.” (A distinct click from the tongue on the palate – tongue_click – as she abruptly turns away, attempting to mask her disappointment with a veneer of indifference.) ``` **Puck** · en (speaking with a US accent) · emotion: shyness/repulsion · bursts: pain_gasp, throat_cough ``` Somewhere, a young adult female, perceived to be in her early twenties, with a low, slightly breathy voice and a US accent, exhibiting moderate expressiveness and confidence, delivering this high-quality studio voice recording with no background noise. (Hesitantly adjusts her posture, a flicker of shyness crossing her face) "I… I’ve been trying needlepoint. Sounds awful, I know." (A resigned_sigh escapes her lips as she anticipates judgment) [Sharp, reflexive inhale triggered by pain.] "Pricked my finger again." (She winces slightly, clutching her finger briefly) "It’s just… quieter than people." (A brief pause, a single, restrained throat_cough to reset speech.) [Single, restrained cough to reset speech.]. "And cleaner than, you know, sculpting. Less messy." (A small, almost self-deprecating chuckle escapes her as she finishes, a hint of repulsion coloring her tone). ``` **Zephyr** · en (speaking with a US accent) · emotion: pleasure/fulfillment · bursts: effort_groan, gag ``` An adult female, appearing to be in her early thirties, with a low, expressive voice and a distinct US accent, tinged with a Valley girl inflection, conveying intense bitterness, resentment, cynicism, and rancor, delivering this high-quality studio voice recording with no background noise. (Slightly hunched shoulders, a cynical smile playing on her lips, she begins with a dismissive wave of the hand) "Okay, you know, optics… optics always just fascinated me." (Her voice gains a touch of wonder as she continues, eyes widening slightly) "Seeing the world explained that way. Like, understanding how light *actually* works, it’s…" (She pauses, a brief effort_groan escaping her as if struggling with a heavy thought) “…ugh, it's just so satisfying.” (She straightens her posture, a flicker of genuine pleasure crossing her face) “Makes you feel like you *get* something fundamental, you know?" (A reflexive choke-like sound, a gag, momentarily interrupts her flow as a suppressed frustration bubbles to the surface.) "And when you finally grasp a difficult concept?" (She takes a deep breath, a visible release, a sigh of relief escaping) "Just...whew...that’s a really good." (She concludes with a small, almost imperceptible nod, the cynicism returning to her expression.) ``` **Despina** · en (speaking with a US accent) · emotion: courage/disappointment · bursts: pleading_whine, hiccup_sob ``` From the first breath, a young adult woman, approximately 26 years old, with a low, expressive voice and a US accent, exhibiting moderate confidence and high arousal, delivering this high-quality studio voice recording with no background noise. (She slumps slightly, shoulders drooping with exhaustion, eyes downcast. A long, shaky inhale.) “Okay, so… obviously, Oxford didn’t pan out.” (A pleading_whine – Upward-tilting, imploring whine seeking favor – escapes her lips, barely audible.) “I thought… I thought I’d really nailed the interview.” (She looks up briefly, a flicker of hope quickly fading.) “I mean, I studied so hard.” (Her voice cracks slightly, a visible tremor in her face. A pause, filled with a deep, frustrated sigh.) “It’s just… frustrating.” (She clutches her chest, a wave of disappointment washing over her. A hiccup_sob – Alternating sobs and diaphragm hiccups – wracks her body, suppressed but audible.) “I just need a minute.” (She closes her eyes, fighting back tears. A long, shuddering exhale.) “I'll apply again next year. Maybe.” (Her voice is barely a whisper, laced with resignation. She opens her eyes, a glimmer of fragile courage returning, but the disappointment lingers.) ``` ### FOCUSED (no demographics — voice quality + delivery only) **Fenrir** · en (speaking with a US accent) · emotion: rapture/fear · bursts: slow_breathing, knuckle_rap ``` Right away, A voice possessing a rich, slightly breathy timbre, colored with thoughtful contemplation and a touch of underlying vulnerability, delivering this high-quality studio voice recording with no background noise. (Standing, gazing intently forward, a subtle tremor in the shoulders suggesting a contained fear. Slow, controlled breaths – *over* there, past the churning waves…) [Paced, controlled breaths for calm.]… (A slight widening of the eyes, a hint of awe creeping in) the horizon just *devours* you whole. (A pause, a fleeting image of vastness) Makes you feel…tiny. (A quick, almost nervous shift of weight) And… vulnerable. [Quick knuckle taps on a hard surface.] (A beat of silence, a searching look) What if the boat fails? (A slight intake of breath, a flicker of genuine surprise) A little bit breathtaking, isn't it? ``` **Aoede** · en (speaking with a US accent) · emotion: attention/exhaustion · bursts: sharp_inhale, fearful_gasp ``` Consider a voice possessing a warm, slightly husky timbre, delivering a tone of weary amusement and reflective self-deprecation, tinged with underlying exhaustion, delivering this high-quality studio voice recording with no background noise. (Standing in a worn kitchen, sunlight filtering through a window, a resigned_sigh escapes as she looks out.) “Okay, another day.” (She pauses, a flicker of amusement crossing her face.) (A beat. She turns slowly, her shoulders slumping slightly, a drone_tone momentarily coloring her voice.) "Knowing this land, you get used to the constant grind. Always something needs doing." (Her gaze drifts to an unseen task.) “Kids fed, cows milked, fields tended…” (A Fast, audible breath signaling realization. – a sharp_inhale) “I forgot to lock the chicken coop last night.” (Her expression shifts, a brief flash of annoyance.) “Great. Wolves, probably.” (A fearful_gasp, a sudden intake of breath colored by anxiety.) “Just… endless.” ``` **Leda** · en (speaking with a US accent) · emotion: malice/carnal desire · bursts: ahem, tongue_click ``` A voice with a slightly dry timbre, tinged with weary bitterness and a subtle undercurrent of malice, delivering this high-quality studio voice recording with no background noise. (Takes a slow, deliberate breath, a flicker of annoyance crossing the face. Pauses, gathering thoughts with a hint of cynicism.) “Quite a few of the undergrads... well, they *think* they're so enlightened now after one philosophy course.” (Ahem — Polite throat-clear to gain attention, a subtle attempt to command attention, laced with condescension.) “It's almost...insulting, really, the naivete.” (A beat of silence, eyes narrowing slightly, a suppressed scoff.) “And the way they look at you sometimes, assessing.” (Tongue_click — Distinct click from tongue on palate, a dismissive gesture, almost imperceptible.) “Like you're just some aged artifact to be cataloged.” (A long, drawn-out pause, a hint of carnal desire momentarily flashing in the eyes, quickly masked by bitterness.) “It.” ``` **Orus** · en (speaking with a US accent) · emotion: scorn/pessimism · bursts: air_kiss, snicker ``` We hear a voice, resonant and dry, tinged with cynicism and a distinct air of disdain, delivering this high-quality studio voice recording with no background noise. (Slight head shake, a dismissive gesture) “Quite frankly, the whole thing's a joke.” (A brief pause, a tightening of the lips) “Another overpaid athlete crying foul.” (A flamboyant, mocking air_kiss gesture with a vocalized 'mwah' but no contact.) “Honestly, I'd sooner watch paint dry.” (A sardonic lift of the eyebrows) “They call it 'sport.’” (A subtle sneer) “More like performance art for simpletons.” (A quick, suppressed snicker – conveying sarcasm and amusement at the absurdity.) “Yeah, they’ll be champions alright,” (A weary sigh, a drop in energy) “of disappointment.” ``` **Rasalgethi** · en (speaking with a US accent) · emotion: peacefulness/trust · bursts: bored_sigh, sharp_inhale ``` We hear a voice, warm and contemplative, with a gentle timbre and a subtly vulnerable tone, delivering this high-quality studio voice recording with no background noise. (A slight pause, head tilted down in thought. A slow, reflective gaze.) “Xavier always said actions spoke louder than words.” (A brief, almost imperceptible shrug.) “He was right. I think… I think I finally trust he *means* well, even when he's being… difficult.” [Bored_sigh] (A long, drawn-out exhale, a flicker of weariness crossing the face.) “And maybe, just maybe, that’s enough.” (Eyes widen slightly, a subtle shift in posture.) He *is* trying. [Sharp_inhale] (A quick, sharp intake of breath, a dawning realization.) “I feel… settled.” (A small, contented smile begins to form.) “Yeah, peaceful.” (The smile deepens, a sense of quiet resolution settling in.) ```