Rhino Speech-to-Intent
Microcontroller Quick Start

Platforms

• Arm Cortex-M4

Requirement

• C99-compatible compiler

Picovoice Account & AccessKey

Signup or Login to Picovoice Console to get your AccessKey. Make sure to keep your AccessKey secret.

Quick Start

Setup

1. Clone the repository:

git clone --recurse-submodules https://github.com/Picovoice/rhino.git

Usage

1. Include the public header files.
2. Link the project to an appropriate library file.
3. Construct the rhino object:

#define MEMORY_BUFFER_SIZE ${MEMORY_BUFFER_SIZE}
static const char* ACCESS_KEY = "${ACCESS_KEY}";

uint8_t memory_buffer[MEMORY_BUFFER_SIZE] __attribute__((aligned(16)));

const uint8_t context_array[] = ${CONTEXT_ARRAY};

const float sensitivity = 0.5f;
const float endpoint_duration_sec = 1.0f;
const bool require_endpoint = true;

pv_rhino_t *rhino = NULL;

const pv_status_t status = pv_rhino_init(
        ACCESS_KEY,
        memory_buffer,
        MEMORY_BUFFER_SIZE,
        context_array,
        sizeof(context_array),
        sensitivity,
        endpoint_duration_sec,
        require_endpoint,
        &rhino);

if (status != PV_STATUS_SUCCESS) {
    // error handling logic
}

4. Pass in frames of audio to the pv_rhino_process function:

extern const int16_t *get_next_audio_frame(void);

while (true) {
    const int16_t *pcm = get_next_audio_frame();
    bool is_finalized = false;
    const pv_status_t status = pv_rhino_process(rhino, pcm, &is_finalized);
    if (status != PV_STATUS_SUCCESS) {
        // error handling logic
    }
    if (is_finalized) {
        // inference event logic/callback
    }
}

5. Release resources explicitly when done with Rhino:

pv_rhino_delete(rhino);

Create Custom Contexts

1. Obtain the UUID of the chipset.
2. Go to Picovoice Console to create models for Rhino Speech-to-Intent engine.
3. Select Arm Cortex-M as the platform when training the model.
4. Select appropriate board type.
5. Train your models.
6. Download your custom context model.
7. Decompress the zip file. The .rhn file inside is the binary model for the Rhino context. The zip archive also contains a .h header file containing the C array version of the binary model.
8. Copy the contents of the array inside the .h header file and update the context_array.

Non-English Languages

Use the corresponding library file (.a) to process non-English contexts. The library files for all supported languages are available on the Rhino Speech-to-Intent GitHub repository.

Demo

For the Rhino Speech-to-Intent mcu SDK, we offer demo projects for several evaluation boards to demonstrate how to use the Rhino Speech-to-Intent engine on microcontrollers. The full list of supported boards are available on the Rhino GitHub repository.

Setup

Clone the repository:

git clone --recurse-submodules https://github.com/Picovoice/rhino.git

Usage

• Arduino Nano 33 BLE Sense (Arm Cortex-M4)
• STM32F411E-DISCO (Arm Cortex-M4)

Resources