The SU03T and SU03T1 are lowcost, lowpower offline speech recognition modules designed for voice-controlled smart home devices, appliances, toys, and industrial automation. Operating completely without internet, they protect user privacy and provide fast, reliable voice interaction.
| Parameter | Value |
|---|---|
| Core | 32-bit RISC at 240MHz, DSP + FPU |
| Memory | 242KB SRAM + 2MB Flash |
| Audio | 1x analog mic input, mono audio output |
| Interfaces | UART, I2C, SPI, PWM, GPIO, 12-bit ADC |
| Power | 5V input, builtin 5Vto-3.3V LDO |
| Speaker Output | Up to 1.6W (8Ω) / 2.4W (4Ω) |
| Standby Current | Approximately 60mA |
+High Performance Processing - 240MHz RISC core with DSP and FPU
+FFT Accelerator Supports 1024point complex FFT for audio processing
+Rich Peripherals - 10 GPIOs, SPI Master/Slave, I2C, 2 PWM outputs
+Complete SDK RTOSbased SDK with UART and I2C control protocol support
√ Compatible with Arduino, ESP32, STM32 via UART
√ Custom voice command support via Smart Era platform
√ SMD package for automated assembly
| Pin | Name | Function |
|---|---|---|
| 1 | VCC | 5V Power Input |
| 2 | DGND | Ground |
| 3 | 3V3 | 3.3V Output |
| 8 | MIC- | ECM Microphone Negative |
| 9 | MIC+ | ECM Microphone Positive |
| 14 | B0 | UARTTXD / I2CSCL |
| 15 | B1 | UARTRXD / I2CSDA |
| 17 | SPK- | Speaker Negative |
| 18 | SPK+ | Speaker Positive |
Module not responding:
Poor voice recognition: