Analyze the attached audio directly and separate the substantive human speakers.
For each speaker whose usable speech supports a complete profile, provide a stable label beginning with Speaker 1 in first-appearance order, primary language, closest dialect or accent, estimated usable speech duration, and a score for every feature below. Briefly identify any additional substantive speakers whose available speech does not support a complete profile. Present the analysis in the format you find clearest.
Every feature is scored from 0 to 100, normalized over the general population. Judge each feature independently. Base the scores on observable speech in this recording. Interpret language-dependent features naturally within the speaker’s primary language.
1. filler_form_preference
0: Filled pauses and discourse holding are almost exclusively non-lexical sounds such as “uh” and “um.”
100: They are almost exclusively lexical fillers or discourse markers such as “like,” “you know,” and equivalents in the speaker’s primary language.
2. syntactic_complexity
0: Speech is constructed from short, simple, linear clauses.
100: Speech uses densely nested, structurally complex multi-clause constructions.
3. clause_completion_habit
0: Clauses frequently fade out, stop mid-thought, or remain unfinished.
100: Clauses consistently reach complete, definite closures.
4. declarative_terminal_rise
0: Declarative clauses almost always end with clearly falling pitch.
100: Declarative clauses almost always end with clearly high-rising pitch.
5. pragmatic_hedging
0: Expression relies on blunt, unmitigated assertions.
100: Expression is pervasively softened, qualified, or hedged.
6. speech_burst_contrast
0: Spoken runs have highly uniform rate and duration with evenly spaced pauses.
100: Speech repeatedly alternates fast, dense word bursts with salient pauses.
7. vocalized_hesitation_prominence
0: Audible fillers and stretched syllables are almost absent during speech planning.
100: Speech planning is pervasively carried by sustained fillers and vowel elongation.
8. pitch_expressiveness
0: Pitch contours show very little variation across phrases.
100: Pitch contours are highly varied, melodic, and animated.
9. lexical_formality
0: Word choice and register are highly casual, colloquial, or slang-heavy.
100: Word choice and register are highly formal, academic, or ceremonious.
10. vocal_gender_presentation
0: Strongly feminine-sounding vocal presentation.
100: Strongly masculine-sounding vocal presentation.
11. perceived_vocal_age_percentile
0: Youngest-sounding vocal extreme in the general population.
100: Oldest-sounding vocal extreme in the general population.
12. pitch_level_percentile
0: Lowest habitual speaking-pitch extreme in the general population.
100: Highest habitual speaking-pitch extreme in the general population.
13. vocal_weight
0: Very light, thin, or reedy vocal substance.
100: Very heavy, thick, or booming vocal substance.
14. accent_markedness
0: Pronunciation is acoustically close to a broadly understood reference variety of the primary language.
100: Pronunciation is extremely regionally or non-natively marked relative to that reference variety.
15. resonance_brightness
0: Very dark, low-frequency-dominant, chest-oriented resonance.
100: Very bright, high-frequency-forward, head-oriented resonance.
16. articulatory_precision
0: Speech is heavily mumbled, slurred, or reduced.
100: Speech is exceptionally crisp, precise, and hyper-enunciated.
17. rhoticity_level
0: R realizations are consistently absent, deleted, or vocalized.
100: R realizations are consistently prominent, hard, or strongly trilled.
18. loudness_dynamic_range
0: Prominent and reduced speech remain at nearly the same loudness.
100: Speech uses exceptionally strong loudness contrast between prominent and reduced material.
19. articulation_tempo
0: Articulated syllables within spoken runs move at an extremely slow rate.
100: Articulated syllables within spoken runs move at an extremely fast rate.
20. vocal_fry
0: Phonation remains consistently clean and modal.
100: Low-register creak or vocal fry is pervasive across voiced speech.
21. modal_breathiness
0: Ordinary voiced speech has firm closure and very little audible air leakage.
100: Ordinary voiced speech has pervasive airy leakage and strongly breathy phonation.
22. modal_roughness
0: Ordinary modal-register voicing has an exceptionally smooth, regular texture.
100: Ordinary modal-register voicing has a pervasive irregular, gravelly texture.
23. vocal_attack
0: Speech onsets usually emerge gradually and softly.
100: Speech onsets usually begin with abrupt, hard glottal closure.
24. sibilant_sharpness
0: Sibilants are extremely dull, retracted, diffuse, or lisped.
100: Sibilants are extremely piercing, dental, focused, or whistling.