Shenzhen Tendzone Intelligent Technology Co., Ltd. is one of the most reliable manufacturers and suppliers of ai audio processor in China. Please feel free to wholesale high quality ai audio processor made in China here from our factory. We also accept customized orders.
AI Audio Processor
Agate V1616 - 16×16 AI Network Audio Processor · AI Denoise · AI-AGC · AEC / ANS / AFC · VaiLink & RTSP
Product Overview
The Agate series uses a domestic ARM chip as its audio processing and management core, delivering professional audio processing and ultra-low-latency, lossless network audio transmission. With built-in AI audio processing, it provides intelligent sound for smart conference rooms, digital courtrooms, multi-function halls, command centers and more - making speech clear and clean, and elevating every voice-communication experience.
Its B/S-architecture software platform lets PCs and mobile devices log in remotely through a browser to adjust processor parameters, retrieve manuals and manage scenes - fitting naturally into any office environment. A 1.8-inch color LCD on the front panel shows the device name, IP address, input/output levels, mute status, current preset and firmware version at a glance.
Key Features
Audio I/O & Network Audio
- 16 channels of balanced microphone/line inputs with 48 V phantom power soft switch.
- 16 channels of balanced line outputs, plus 16-channel RTSP network audio streaming.
- 16 × 16 VaiLink network audio - lossless, ultra-low-latency audio over IP.
- Optimized preamp gain with 8 adjustable steps (0 / -6 / -12 / -18 / -24 / -30 / -36 / -48 dB), accommodating both MIC and LINE levels.
- USB 2.0 audio interface for recording and playback via USB flash drive.

AI & Acoustic Processing
- Built-in AI audio algorithms: AI noise reduction removes both stationary and non-stationary noise - keyboard typing, mouse clicks, chair movement, door slamming, office machines, road traffic, footsteps and more.
- AI-AGC uses intelligent recognition to automatically bring loudness to the preset target level, replacing frequent manual adjustment.
- 2 independent adaptive Acoustic Echo Cancellation (AEC) modules: selectable tail length of 128 / 256 / 512 ms, 80 dB/s convergence, up to 80 dB echo reduction with 3 adjustable levels.
- 2 independent adaptive Noise Suppression (ANS) modules with 4 levels - 9 / 12 / 15 / 18 dB.

AI Audio Processing Engine
Every input channel flows through a full chain of intelligent processing - AI denoise, adaptive noise suppression, adaptive echo cancellation, adaptive feedback cancellation, AGC / AI-AGC, auto-mixing and parametric EQ. Two independent AEC modules and two independent ANS modules run in parallel, so hybrid meetings with multiple far-end sources stay echo-free, while the embedded NPU provides dedicated AI computing power on-device.
Application Scenarios
- Smart conference rooms - AI denoise and AEC keep every remote participant clearly heard.
- Digital courtrooms - intelligible, stable speech reinforcement for proceedings.
- Multi-function halls - fast preset switching and zone-based Mix-Minus reinforcement.
- Command centers - reliable, always-on intelligent audio for mission-critical dispatch.
- Conference centers - 16 × 16 VaiLink and RTSP streaming tie multiple rooms together.
- Training and lecture rooms - USB-B BYOD audio and camera tracking in one box.
|
|

System Solution
Microphones feed the 16 balanced inputs, where the Agate V1616 applies the full AI processing chain. Processed audio is routed to amplifiers and speakers through the 16 balanced outputs, shared losslessly over 16 × 16 VaiLink network audio, streamed as 16 RTSP channels, or sent to video-conferencing PCs and BYOD devices through the USB-B sound card with far-end echo fully cancelled. The built-in camera tracking and 8 GPIO / RS-232 / UDP control tie audio together with cameras, lights, curtains and projection screens, while engineers tune everything remotely from a browser.

About Tendzone
Established in 2010, Tendzone is a global leader in providing advanced audio-visual (AV) solutions and manufacturing high-quality AV products. We specialize in a wide range of cutting-edge technologies, including audio processors, microphones, speakers, power amplifiers, AV over IP systems, digital conference systems, and MIDIS Distributed Multimedia Transmission Control Systems. Our solutions are trusted across industries such as conference rooms, command centers, education, multi-functional halls, and stadiums.

Our Service
Our dedicated customer support team offers reliable technical support, quick response times, and preventive maintenance service.

Pre-Sales
We assess your unique requirements, provide tailored consultations and demonstrations, and design a customized AV system to ensure the perfect solution.

In-Sales
We deliver transparent proposals, manage the full project implementation, and provide initial training to ensure a smooth and successful setup.

After-Sales
We offer prompt technical support, customized maintenance plans, continuous software upgrades, and actively use your feedback to drive ongoing service improvement.
Tendzone, established in 2010, is a global leader in providing advanced audio-visual (AV) solutions and manufacturing high-quality AV products. We specialize in a wide range of cutting-edge technologies, including audio processors, microphones, speakers, power amplifiers, AV over IP systems, digital conference systems, and MIDIS Distributed Multimedia Transmission Control Systems. Our solutions are trusted across industries such as conference rooms, command centers, education, multi-functional halls, and stadiums.
Our address
Building 1A, Building 16, Shenzhen International Innovation Valley, Dashi 1st Road, Nanshan District, Shenzhen
Phone Number
Mobile/Wechat/WhatsApp : (+86) 13302998794
travis.tu@tendzone.com

Hot Tags: ai audio processor, China ai audio processor manufacturers, suppliers, factory, Audio DSP, 8x8 Web Based Audio Processor, 16x16 Web Based Dante Audio Processor, audio dsp processor, audio processor, Audio Signal processing
Technical Specifications
|
Processing Core |
Domestic ARM chip (secure & controllable), integrated real-time OS, embedded NPU for AI computing power |
|
Sampling Rate |
48 kHz, 24-bit |
|
Frequency Response |
20 Hz – 20 kHz (±0.5 dB) |
|
Dynamic Range (A/D/A) |
>110 dB, A-weighted (20 Hz–20 kHz @ 0 dB gain) |
|
THD+N |
≤0.005% @ 1 kHz, +4 dBu |
|
Noise Floor (A-weighted) |
≤-95 dBu |
|
Channel Isolation |
≤-105 dB @ +20 dBu, 1 kHz |
|
Maximum Input Gain |
48 dB (8-step preamp: 0 to -48 dB) |
|
Maximum Output Level |
20 dBu |
|
Input Impedance |
2 kΩ balanced |
|
Output Impedance |
200 Ω balanced |
|
Phantom Power |
DC 48±6 V, 10 mA |
|
AEC |
2 independent modules; tail 128 / 256 / 512 ms selectable; convergence 80 dB/s; echo reduction up to 80 dB; 3 effect levels |
|
ANS |
2 independent modules; noise reduction 4 levels - 9 / 12 / 15 / 18 dB |
|
AI Denoise |
Removes stationary and non-stationary noise (keyboard, mouse clicks, chair movement, doors, office machines, traffic, footsteps) |
|
AI-AGC |
Intelligent target-level automatic gain |
|
AFC |
Per channel; 16-point notch AFC or frequency-shift AFC |
|
Input Channel Processing |
Preamplifier, signal generator, expander, compressor, AFC, 10-band parametric EQ (PEQ / High-Shelf / Low-Shelf / LP / HP) |
|
Output Channel Processing |
Selectable 10 / 15 / 31-band parametric EQ (PEQ / High-Shelf / Low-Shelf / NOTCH), delay, HPF / LPF, limiter |
|
Network Audio |
16 × 16 VaiLink network audio; 16-channel RTSP streaming output |
|
USB |
USB 2.0 (flash-drive record & playback); USB-B sound card 2 × 2 (Windows / iOS / Android / HarmonyOS) |
|
Control |
B/S web UI (Chinese / English, customizable UI), UDP, RS-232, open third-party protocol |
|
GPIO |
8 × configurable (input or output, software-set) |
|
Display |
1.8" color LCD (device name, IP, I/O levels, mute, preset, firmware) |
|
Presets |
50 groups |
|
Power Supply |
AC110-220V ±10%, 50/60 Hz |
|
Power Consumption |
36 W |
|
Dimensions (W × D × H) |
484 × 280 × 44 mm (1U, standard 19" rack); package 570 × 410 × 135 mm |
|
Weight |
3.1 kg net; 3.6 kg gross |

















