Speech production reduces to a source-filter model with glottal or noise sources
Evidence
In Figure 5.11, speech production is finally reduced into a signal model, the so-called source–filter model. In voiced signals (in phonation), the source is the glottal oscillation and in unvoiced sounds the source is frication noise created in a constriction or a transient sound generated by rapid release of pressure when a vocal tract closure is opened. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 600
Categorization
Category: Definitional Content
- DEFINE: Defines the canonical source-filter model and how unvoiced sounds are generated.
- XREF: Source-filter decomposition is foundational to speech synthesis, coding, and vocoder applications — directly serves the reader’s application goal.
Related
- Voiced and unvoiced speech differ in whether source is glottal oscillation or frication noise (similarity 0.94)
- Speech coding exploits production models for low bit rates without databases (similarity 0.91)
- The vocal apparatus functions as a pressure source driving a resonant filter system (similarity 0.87)
- Speech production reduces to lungs as compressor driving a glottal oscillator through resonating tracts (similarity 0.87)
- A unified source-filter model describes human and synthetic voice alike (similarity 0.85)
- A huge variety of speech sounds can be made without vocal folds (similarity 0.84)
- Mathematical source-filter models of speech enable practical synthesis and coding algorithms (similarity 0.84)
- The speech tract is modeled as a transmission line with pressure and volume-velocity variables (similarity 0.83)
- The singing voice reduces to a source plus a shapeable vocal-tract filter (similarity 0.83)
- The glottal waveform’s closing phase is abrupt and carries most excitatory energy. (similarity 0.83)
- Glottal function is modeled as mass-spring or distributed transmission-line systems (similarity 0.83)
- Singing is modeled as vocal-fold excitation filtered by a time-varying vocal tract (similarity 0.83)
- All musical instruments and voices share a source-filter acoustic structure (similarity 0.82)
- Musical instruments couple excitation to resonances; human speech does not (similarity 0.81)
- The vocal tract can be simulated as a cascade of acoustic transmission line segments (similarity 0.81)