WTK6900FC-M01 Self-Learning Voice Recognition Module
5–8m Far‑Field Recognition with 20 Customizable APP Commands & 150 Fixed Commands
The WTK6900FC-M01 Self-Learning Voice Recognition Module is a self‑learning voice recognition module featuring 5–8 meter far‑field voice capture and 20 customizable command words (1 wake word + 19 command words) via a dedicated mobile APP. Powered by DNN-HMM deep learning technology, it delivers up to 98% recognition accuracy in quiet environments. With 150 fixed commands, UART control, and support for Chinese, English, Japanese, and Korean, this module is the ideal solution for smart home appliances, industrial controls, and customizable voice interaction products.
- Self‑learning – customize wake words & commands via APP
- 5–8m far‑field – robust voice pickup
- 20 custom commands – 1 wake + 19 commands
- 150 fixed commands – pre‑loaded for instant use
- Multi‑language – CN, EN, JP, KR
Self‑Learning Feature
Wake Word Learning
- Customize the wake word via APP
- Record or upload audio (max 8 seconds)
- Supports 1 custom wake word
Command Word Learning
- Up to 19 custom command words
- Commands ≤ 5 Chinese characters
- Learn via APP (recording or file upload)
TTS Reply Customization
- Customize voice responses via TTS
- Multi‑language TTS (CN, EN, JP, KR, etc.)
- Adjustable volume, pitch, and speed
Delete Learned Commands
- On‑board DEL key for quick deletion
- Delete individual commands via APP
- LED indicators for learning/deletion status
Self-Learning Voice Recognition Module Key Features
On‑Board Controls & Indicators
| Control | Function |
|---|---|
| Learn Key (Short) | Play previous command’s reply voice |
| Learn Key (Long) | Enter self‑learning mode for the current command |
| Delete Key (Short) | Play next command’s reply voice |
| Delete Key (Long) | Delete the learned data for the current command |
| BLE LED | Bluetooth connection success |
| DEL LED | Entering delete mode for learned voice |
| LEARN LED | Entering self‑learning mode |
| POWER LED | Power supply normal |
Technical Specifications
| Parameter | Value / Condition |
|---|---|
| Module Type | Self‑learning voice recognition module |
| Recognition Technology | DNN-HMM deep neural network |
| Far‑Field Range | 5 – 8 meters |
| Fixed Commands | 150 pre‑loaded commands |
| Self‑Learning Commands | 20 (1 wake word + 19 command words) |
| Recognition Accuracy | Up to 98% (quiet environment) |
| Languages Supported | Chinese, English, Japanese, Korean |
| Audio Output | 8Ω / 1W (standard), 4Ω / 3W (optional) |
| Audio Format | MP3, 16k–24kbps |
| Microphone Input | ECM (electret condenser microphone) |
| Recommended MIC | 4015 microphone, -36dB, 2.0 terminal, 150mm cable |
| Control Interface | UART (3.3V TTL) |
| Operating Voltage | 3.5V – 5.0V (Type-C 5V / 1.2A) |
| Working Current | ≤ 1.2A (active), ≤ 40mA (standby) |
| Charging Current | ≤ 300mA (default, for Li‑ion version) |
| Flash Memory | 32MB |
| Package | SSOP24 |
| Operating Temperature | ‑40°C ~ +85°C |
Command Output Examples
| Command | UART Output |
|---|---|
| Wake Word: “你好小唯” (Hello Xiaowei) | 7E 06 FF 06 01 01 0D EF |
| “打开空调” (Turn on AC) | 7E 06 FF 06 01 02 0E EF |
| “关闭空调” (Turn off AC) | 7E 06 FF 06 01 03 0F EF |
| “增大风速” (Increase fan speed) | 7E 06 FF 06 01 04 10 EF |
| “减小风速” (Decrease fan speed) | 7E 06 FF 06 01 05 11 EF |
| “升高一度” (Increase temperature) | 7E 06 FF 06 01 06 12 EF |
| “降低一度” (Decrease temperature) | 7E 06 FF 06 01 07 13 EF |
| “全自动” (Full auto) | 7E 06 FF 06 01 08 14 EF |
Protocol: 7E + Length + Command + Parameter + Checksum + EF. Custom self‑learned commands output unique IDs.
Voice Recognition Selection Guide
| Model | Recognition Range | Commands | Self‑Learning | Languages | Power |
|---|---|---|---|---|---|
| WTK6900FC-M01 | 5–8m | 150 fixed + 20 self‑learn | ✓ (APP) | CN, EN, JP, KR | 50–60mA |
| WTK6900HC | 5–8m | 300 fixed + 1 wake self‑learn | Wake only | CN, EN | 25–30mA |
| WTK6900HA | 3–5m | 300 fixed | ✘ | CN, EN | 15–20mA |
| WTK6900L | 0.5–3m | 20 fixed | ✘ | CN | 20–25mA |
| WT2606A | 3–5m | 10 wake + 300 commands | ✘ | CN, EN | 30–40mA |
WTK6900FC-M01 offers the most flexible self‑learning capability with APP‑based customization.
Key Interfaces
| Interface | Function |
|---|---|
| Type-C | Power supply (5V / 1.2A) + charging |
| UART (VCC, GND, TX, RX) | Serial communication (3.3V TTL) |
| MIC (2.0 terminal) | External microphone connection (check polarity) |
| Speaker (2.0 terminal) | 8Ω 1W speaker output (standard) |
| Upgrade Port (GND, DP, DM, VCC) | Firmware programming |
| Battery | Li‑ion battery charging (optional) |
Application Scenarios
Why Choose WTK6900FC-M01 Self-Learning Voice Recognition Module
Frequently Asked Questions
What is the WTK6900FC-M01 Self-Learning Voice Recognition Module?
WTK6900FC-M01 Self-Learning Voice Recognition Module is a self-learning voice recognition module with 5–8m far-field recognition, 20 customizable command words via APP (1 wake word + 19 command words), 150 fixed commands, and UART control.
What is Self-Learning Voice Recognition Module?
Self-Learning Voice Recognition Module allows users to customize wake words and command words via the mobile APP. You can record or upload audio files to create personalized voice commands without needing to reprogram the module.
How many custom commands can be learned?
The Self-Learning Voice Recognition Module supports up to 20 self-learning command words, including 1 wake word and 19 command words. Command words should not exceed 5 Chinese characters.
What is the recognition range?
Self-Learning Voice Recognition Module supports 5–8 meter far-field voice recognition, making it suitable for room-scale voice control applications.
How is the module controlled?
It uses a standard UART interface (3.3V TTL) for communication. The module outputs command IDs via UART when voice commands are recognized.
What languages are supported?
Self-Learning Voice Recognition Module supports Chinese, English, Japanese, and Korean for voice recognition and TTS output.
What APP is used for self-learning?
A dedicated APP (available for iOS and Android) is used to add wake words, command words, and TTS responses. The APP supports recording, file upload, and TTS text input.
What is the recognition accuracy?
In quiet environments, the recognition accuracy reaches up to 98% for fixed commands and good performance for self-learned commands.
What microphone is recommended?
The module supports electret microphones (ECM). Recommended model: JMO-627BA283R-10TC38 (4015 microphone, -36dB, 2.0 terminal, 150mm cable).
What are the typical applications?
Smart home appliances, smart lighting, smart bathrooms, smart furniture, smart toys, industrial controls, automotive voice systems, and any application requiring customizable voice control.
Important: Self‑learning must be performed in a quiet environment with normal speaking volume within 0.5 meters. Command words should not exceed 5 Chinese characters. If learning fails multiple times, the first recording may contain noise – re‑learn the command. The APP connection range is within 5 meters. The module supports both iOS and Android devices. For custom firmware or command set modifications, contact Waytronic support.
