Skip to main content

Text-to-Speech Example

This example demonstrates how to programmatically send text-to-speech (TTS) commands to your Aria Gen2 device using the Python SDK. You can use TTS to provide audio feedback, instructions, or notifications to the device wearer.

Quick Start​

Run the text-to-speech example script:

python ~/Downloads/projectaria_client_sdk_samples_gen2/device_tts.py --text "Hello from Aria"

The device will speak the text through its built-in speakers.

What This Example Does​

The script performs the following operations:

  1. Connects to the device
  2. Sends a text-to-speech command with your specified text
  3. The device converts the text to speech and plays it through the speakers
  4. Returns once the command is sent (device plays audio asynchronously)

Code Walkthrough​

Step 1: Import Required Modules​

import argparse
import aria.sdk_gen2 as sdk_gen2

Step 2: Parse Command-Line Arguments​

def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--text",
        dest="text",
        type=str,
        default="",
        required=True,
        help="TTS text to rendered by the device.",
    )
    return parser.parse_args()

Required Argument:

  • --text: The text string you want the device to speak

Step 3: Connect to Device​

device_client = sdk_gen2.DeviceClient()
config = sdk_gen2.DeviceClientConfig()
device_client.set_client_config(config)

try:
    device = device_client.connect()
    print(f"Connected to device {device.connection_id()}")
except Exception as e:
    print(f"Failed to connect: {e}")
    return

This establishes a connection to the first available device.

Step 4: Send Text-to-Speech Command​

text = args.text
print(f"Sending text-to-speech: '{text}'")
device.render_tts(text)
print("Text-to-speech command sent successfully")

How it works:

  • The render_tts(text) method sends the text to the device
  • The device processes the text and generates speech
  • Audio is played through the device's built-in speakers
  • The function returns immediately (audio plays asynchronously)

Complete Example Code​

Here's the full text-to-speech script:

import argparse
import aria.sdk_gen2 as sdk_gen2


def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--text",
        dest="text",
        type=str,
        default="",
        required=True,
        help="TTS text to rendered by the device.",
    )
    return parser.parse_args()


if __name__ == "__main__":
    args = parse_args()

    # Set up the device client to initiate connection to the device
    device_client = sdk_gen2.DeviceClient()
    # Set up the device client config to specify the device to be connected to e.g. device serial number.
    # If nothing is specified, the first device in the list of connected devices will be connected to
    config = sdk_gen2.DeviceClientConfig()

    device_client.set_client_config(config)
    device = device_client.connect()
    print(f"Connected to device: {device.connection_id()}")
    print(f"Rendering TTS: {args.text}")
    device.render_tts(text=args.text)

Usage Examples​

Simple Message​

python ~/Downloads/projectaria_client_sdk_samples_gen2/device_tts.py --text "Recording started"

Multi-Word Message​

python ~/Downloads/projectaria_client_sdk_samples_gen2/device_tts.py --text "Please look at the target"

Instructions​

python ~/Downloads/projectaria_client_sdk_samples_gen2/device_tts.py --text "Turn left at the next intersection"

Numbers and Punctuation​

python ~/Downloads/projectaria_client_sdk_samples_gen2/device_tts.py --text "You have 3 notifications. Check your device."

The TTS engine handles numbers and punctuation naturally.

Troubleshooting​

TTS Command Sent But No Audio​

Problem: The render_tts() command succeeds but device doesn't speak.

Solutions:

  1. Check device volume:

    • Ensure device audio is not muted
    • Check volume level in Companion App

  2. Check device status:

    • Verify device is powered on and functioning
    • Check battery level (low battery may affect audio)

  3. Verify connection:

    device = device_client.connect()
    print(f"Connected: {device.connection_id()}")

  4. Test with simple message:

    device.render_tts("Test")

Connection Lost During TTS​

Problem: Device disconnects when sending TTS.

Solutions:

  1. Verify USB connection is stable
  2. Check authentication: 
    aria_gen2 auth check
  3. Reconnect if needed: 
    try:
        device.render_tts(text)
    except Exception as e:
        print(f"Connection lost: {e}")
        device = device_client.connect()  # Reconnect
        device.render_tts(text)  # Retry

Next Steps​