📄️ AEC Node - Echo Cancellation
AEC node reference for Switchboard Audio SDK. Remove echo from voice calls, conferencing, and voice AI apps. Supports WebRTC AEC3 on iOS, Android, and desktop.
📄️ Noise Filter
A noise filter node reduces unwanted background noise from an audio stream, improving clarity in recordings and live communications. It is especially useful in environments with consistent ambient sounds like hums, hisses, or background chatter. Common features include adjustable noise reduction levels, real-time processing, and optional voice activity detection (VAD) for further enhancement.
📄️ Real-Time Communication (RTC)
An RTC (Real-Time Communication) node enables low-latency, two-way audio transmission over a network, supporting use cases like voice chat, conferencing, and live collaboration. It typically handles audio encoding/decoding, jitter buffering, packet loss concealment, and synchronization. Many RTC nodes also support integration with signaling protocols and NAT traversal techniques (e.g., STUN, TURN).
📄️ Speech-to-Text (STT)
A speech-to-text (STT) node transcribes spoken audio into written text, enabling voice control, transcription services, and accessibility features. It typically supports real-time and batch processing, multiple languages, speaker diarization, and punctuation handling. Some implementations also offer confidence scoring and formatting options.
📄️ Text-to-Speech (TTS)
A text-to-speech (TTS) node converts written text into spoken audio, enabling voice-based interaction in applications like virtual assistants, screen readers, and automated announcements. Typical features include support for multiple languages, voice styles (e.g., male, female, neural), adjustable speaking rate, and pitch control.
📄️ VAD Node - Voice Activity Detection
Voice Activity Detector node for Switchboard Audio SDK. Detect speech in audio streams in real time. Supports SileroVAD and built-in detection.