$8 LU-ASR01 offline speech recognition board features “TW-ASR ONE” chip

LU-ASR01 is a board capable of offline speech recognition with a built-in microphone, a speaker connector, twelve through holes for GPIOs and a temperature sensor interface for DHT11/DS18B20,  plus a USB Type-C port for power and programming.

At first, I thought it might be based on the Unisound US516P6 microcontroller which we’ve seen on some inexpensive offline speech recognition modules, but the chip looks completely different, with the marking TW-ASR ONE. So let’s investigate…

LU-ASR01 offline speech recognition moduleLU-ASR01 board specifications took some effort, but here’s what I’ve managed:

  • MCU – TW-ASR ONE (aka TWen ASR ONE) microcontroller with 4MB flash, 512KB RAM, and a BNPU for voice processing; package: QFN48L (6x6x0.85mm)
  • Audio I/O
    • Built-in microphone
    • 2-pin speaker header plus 3W power amplifier for 4Ω/3W speaker
  • Voice recognition
    • Up to 10 meters wake-up range
    • 98% ultra-high recognition rate
    • Customizable to 5 wake-up words and 200 recognition words
  • USB – 1x USB Type-C port for power and programming via CH340C USB to TTL chip
  • Expansion – 12x through holes with 8x GPIOs, of which 6 can be used as PWM, one serial Tx, one DHT11/DS18B20 temperature sensor interface
  • Dimensions – About 3.7 x 3.7 cm
  • Weight – 35g
Finding documentation was a nightmare, and I had to ask Banggood, which provided a link to a RAR file with some documentation and video tutorials in Chinese. You’ll need to download the Blockly-based TwenBlock program for Windows in order to visually program the board.
TWenBlock ASR One visual programming
TWenBlock screenshot

Everything is done within this interface including defining wake words and voice commands, monitoring the temperatures, and controlling I/Os, in the sample above some servos. The program is then built and flashed to the board over USB. It’s also possible to directly modify the source code generated from Blockly. At this point, I don’t know if it supports languages other than Mandarin.

If you were to go ahead and give it a try, you may want to check out a 5-part tutorial on Zhihu social network, again in Chinese, that’s more detailed than the information in the RAR file, and in text form, so using machine translation is easier… I could also find a simplified block diagram of the ASR ONE microcontroller as well, and learned that the system runs FreeRTOS.

TWen ASR ONE
TWen ASR ONE block diagram

Note that the review is using another ASR ONE v0.1 board that looks more like a NodeMCU board, but I could not find it for sale.

ASR ONE v0.1 offline speech recognition board

The LU-ASR01 offline speech recognition board is available now and sold on Banggood for $7.99 including shipping, and you’ll also find it on Aliexpress. Some other shops list it as an upgrade to VRM LD3320 board based on LD3320A and usually selling for over $20.

Share this:

Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress

ROCK Pi 4C Plus
Subscribe
Notify of
guest
The comment form collects your name, email and content to allow us keep track of the comments placed on the website. Please read and accept our website Terms and Privacy Policy to post a comment.
7 Comments
oldest
newest
sander
sander
1 year ago

Who do you need to program/configure it? Because it has a limited vocabulary? Just a few words / sentences, like “turn on heating”, “turn off lights”?

Gimpson
1 year ago

I’m about to try these, but not sure if it English commands and wake words are supported. Do you think it supports English commands or wake words?

gimpson kura
1 year ago

Sure but creating an account to gain access to their customizing tool is just impossible. I don’t live in China, and they keep asking for a Chinese phone number for verification

Razo
Razo
1 year ago

it works with English. You can type any English paragraphs into the source code and let the cloud generate the audio library. You then compile the audio lib into the local codes and download the final image to the chip. It gives pretty legit synthesized voice. Yes, u need to be logged in and a Mainland China phone number is needed

Gimpson
1 year ago

Thanks a lot! For my application, I just need the module to output predefined data via UART to a microcontroller, then responses will be handled by a df mini. But just a quick question; do I need a cloud account for customization of command words? My thought is; since it’s AI, it can automatically recognize any spoken words in english and output a predefined data via UART.

Khadas VIM4 SBC