LocalVQE v1.4-AEC
Esta página de Soniqo documenta LocalVQE v1.4-AEC tal como está implementado en speech-swift. Los enlaces a Hugging Face aparecen debajo de las notas de integración.
Primero página interna
Las tarjetas y menús apuntan primero aquí; los enlaces al modelo fuente y a los bundles siguen disponibles en esta página.
Resumen
| Modelo | LocalVQE v1.4-AEC |
|---|---|
| Rol | Explicit-reference acoustic echo cancellation |
| Backend | Native C++ adaptive filter plus stateful Core ML residual mask |
| Salida | 16 kHz mono Float32 clean-microphone PCM |
| Idiomas | Language-independent acoustic processing |
| Licencia | Apache-2.0 code and weights |
| Estado | Streaming and complete-clip speech-swift runtime |
| Fuente | LocalVQE |
| Producto Swift | SpeechEnhancement |
| CLI / runtime | Programmatic LocalVQEEchoCanceller runtime |
Uso
El fragmento siguiente refleja la API o el comando actual expuesto por speech-swift.
import SpeechEnhancement
let aec = try await LocalVQEEchoCanceller.fromPretrained()
let cleanMicrophone = try aec.processFrame(
microphone: microphoneFrame,
reference: playbackReferenceFrame
)Enlaces del modelo
Notas de implementación
- Microphone and playback reference are separate, synchronized inputs; a mixed system-audio transcript is not a valid reference.
- Streaming accepts exactly 256 mono samples per input at 16 kHz and adds one 16 ms spectral hop of algorithmic latency.
- Reset after dropped frames, device changes, capture restarts, or recording boundaries so stale filter and neural state cannot cross discontinuities.
- A silent reference preserves microphone-only speech; the runtime never falls back to subtracting a mix or muting the microphone.
- Do not blindly cascade this path with Apple Voice Processing I/O because two adaptive cancellers can distort near-end speech.