Skip to main content

Running your_chat_bot on DshanPi-A1

This document walks you through running the TuyaOpen your_chat_bot chat bot project on the DshanPi-A1 board.

Prerequisitesโ€‹

Read the Quick Start and its sub-sections to understand:

Build Methodsโ€‹

DshanPi-A1 supports two build methods:

  • Cross-compilation: Build on a PC, then transfer and run on the board
  • Native build: Build directly on the board

The system automatically detects the current platform and selects the appropriate build method.

Note: Cross-compilation is not supported on macOS. Use Linux or build directly on the board.

Quick Start Guideโ€‹

To run on DshanPi-A1, you need to complete two extra steps:

  1. Configure the onboard sound card (so the board can record and play audio)
  2. Configure the voice wake word model path

Step 1: Configure the Onboard Sound Cardโ€‹

The DshanPi-A1 has a built-in microphone and speaker, but they must be configured before use.

1.1 Install Audio Librariesโ€‹

Install the ALSA library (for audio input and output):

sudo apt-get install libasound2-dev

1.2 Configure Speaker Outputโ€‹

The DshanPi-A1 has two audio outputs: the headphone jack and the onboard speaker. For convenience, we configure the onboard speaker as the default output.

Edit the audio configuration file:

sudo vi /etc/asound.conf

Add the following content:

pcm.speaker_r {
  type route
  slave.pcm "hw:0,0"
  slave.channels 2
  ttable.0.1 1
  ttable.1.1 0
  ttable.0.0 0
  ttable.1.0 0
}

Save and exit (press ESC, type :wq, then press Enter).

Step 2: Configure the Voice Wake Word Modelโ€‹

Voice wake word uses a KWS (Keyword Spotting) model so the device can recognize โ€œไฝ ๅฅฝๆถ‚้ธฆโ€ (Hello Tuya).

2.1 Obtain the Model Filesโ€‹

You can get the wake word model in two ways:

Option 1: Automatic download

  • When building for the DshanPi-A1 platform, the model is automatically downloaded to: 
    platform/LINUX/tuyaos_adapter/src/tkl_audio/models

Option 2: Manual download

Use the following commands to download mdtc_chunk_300ms.mnn and tokens.txt to ~/tuyaopen_models:

wget -P ~/tuyaopen_models https://github.com/tuya/TuyaOpen-ubuntu/raw/platform_ubuntu/tuyaos_adapter/src/tkl_audio/models/mdtc_chunk_300ms.mnn
wget -P ~/tuyaopen_models https://github.com/tuya/TuyaOpen-ubuntu/raw/platform_ubuntu/tuyaos_adapter/src/tkl_audio/models/tokens.txt

The default path in the code is ~/tuyaopen_models. If you need to change it, follow the steps below.

2.2 Configure the Model Pathโ€‹

Set the model file paths in the project:

  1. Go to the project directory:

    cd apps/tuya.ai/your_chat_bot

  2. Select the DshanPi-A1 configuration:

    tos.py config choice

    In the menu, choose DshanPi_A1.config

  3. Open the config menu:

    tos.py config menu

  4. Navigate to the options:

    (Top) โ†’ Choice a board โ†’ LINUX โ†’ TKL Board Configuration

  5. Set these two options to the actual paths of the model files:

    • KWS model file path โ†’ full path to mdtc_chunk_300ms.mnn
    • KWS model token file path โ†’ full path to tokens.txt

2.3 Configuration Exampleโ€‹

If the model files are in ~/tuyaopen_models, the configuration should look like this:

models_path_config
models_path_config

Important: The paths must be correct, or the voice wake word feature will not work.

Additional Notesโ€‹

Transferring Files When Using Cross-Compilationโ€‹

If you cross-compile on a PC, you need to copy the built executable to the board. Use scp:

scp ./dist/your_chat_bot_1.0.1/your_chat_bot_QIO_1.0.1.bin username@192.168.1.xxx:~/

Command parameters:

  • username โ€” Username on the board
  • 192.168.1.xxx โ€” IP address of the board
  • ~/ โ€” Target directory on the board

Running the Executableโ€‹

./your_chat_bot_QIO_1.0.1.bin

Command parameters:

  • your_chat_bot_QIO_1.0.1.bin โ€” Executable file name

On the first run, you need to perform device network configuration (pairing). If pairing or network connection fails, try deleting the tuyadb folder and running the program again.

FAQโ€‹

Q: Voice wake word is not working. What should I do?
A: Check that the model file paths are correct and that the files are complete. You can use ls -lh to verify that the files exist and that their sizes are as expected.

Q: The executable cannot be run. What should I do?
A: Check that the executable has execute permission. You can run chmod +x your_chat_bot_QIO_1.0.1.bin to grant execute permission.