ReSpeaker 2-Mics Pi HAT with Raspberry PI

Overview

A ReSpeaker 2-Mics Pi HAT is a dual-microphone expansion board for Raspberry PI. This means you can build a portable project that integrates Amazon Alexa Voice Service, Google Assistance, can be use also as voice recognition, voice recording, and many more

This post is a tutorial on how to use multiple features of the ReSpeaker 2-Mics Pi HAT expansion board. These features are record audio using dual-microphone, using 3.5mm audio jack connected to speaker or headphone to output audio, interface user button, and interface 3 APA102 RGB LEDs.

Hardware

• Raspberry Pi 3B+ (Compatible on: Raspberry Pi Zero and Zero W, Raspberry Pi B+, Raspberry Pi 2 B, Raspberry Pi 3 B, Raspberry Pi 3 B+, Raspberry Pi 3 A+)

Software

• RPI Terminal
• Thonny Python IDE (Pre installed on RPI OS)

Application Discussion

A ReSpeaker 2-Mics Pi HAT is a dual-microphone expansion board for Raspberry PI. The board is developed based on WM8960, is a low power, high quality stereo CODEC designed for portable digital audio applications. 

ReSpeaker 2-Mics Pi Hat Expansion Board:

Other than its audio feature, here are the other components included on the board.

ReSpeaker 2-Mics Pi Hat components:

• Dual Microphones(Left and right microphones)
• WM8960: a low power stereo codec
• POWER: Micro USB port for powering the ReSpeaker 2-Mics Pi HAT, power the board for providing enough current when using the speaker.
• User Button (Connected to RPI GPIO17)
• RGB LED (3 APA102 RGB LEDs, connected to SPI interface)
• 3.5mm Audio Jack (Connecting headphone or speaker with Audio plug)
• Raspberry Pi 40-Pin Headers: support Raspberry Pi Zero, Raspberry Pi 1 B+, Raspberry Pi 2 B , Raspberry Pi 3 B and Raspberry Pi 3 B+
• JST 2.0 Speaker out (Connecting speaker with JST2.0 connector)
• I2C Grove Interface (Connected to RPI I2C-1)
• GPIO12 Grove interface (Connected to RPI GPIO12 & GPIO13)

Set-up the Hardware   

Connect the ReSpeaker 2-Mics Pi HAT to Raspberry PI and connect it like this.

Raspberry PI Connection

Raspberry PI Zero Connection

Set-up the Software

Install ReSpeaker 2-Mics Pi HAT Driver:

• Install the seeed voice card drivers by using RPI Terminal.

sudo apt-get update
sudo apt-get upgrade
git clone https://github.com/respeaker/seeed-voicecard.git
cd seeed-voicecard
sudo ./install.sh
reboot 

• Check that the sound card name matches the source code seeed-voicecard by command aplay -l and arecord -l.

pi@raspberrypi:~ $ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: b1 [bcm2835 HDMI 1], device 0: bcm2835 HDMI 1 [bcm2835 HDMI 1]
  Subdevices: 4/4
  Subdevice #0: subdevice #0
  Subdevice #1: subdevice #1
  Subdevice #2: subdevice #2
  Subdevice #3: subdevice #3
card 1: Headphones [bcm2835 Headphones], device 0: bcm2835 Headphones [bcm2835 Headphones]
  Subdevices: 4/4
  Subdevice #0: subdevice #0
  Subdevice #1: subdevice #1
  Subdevice #2: subdevice #2
  Subdevice #3: subdevice #3
card 2: seeed2micvoicec [seeed-2mic-voicecard], device 0: bcm2835-i2s-wm8960-hifi wm8960-hifi-0 [bcm2835-i2s-wm8960-hifi wm8960-hifi-0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
pi@raspberrypi:~ $ arecord -l
**** List of CAPTURE Hardware Devices ****
card 2: seeed2micvoicec [seeed-2mic-voicecard], device 0: bcm2835-i2s-wm8960-hifi wm8960-hifi-0 [bcm2835-i2s-wm8960-hifi wm8960-hifi-0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
pi@raspberrypi:~ $

• Test speaker and mic by command " arecord -f cd -Dhw:(card #) | aplay -Dhw: (card #)". Replace the card # as corresponding to your device card number as shown as example here "card 2: seeed2micvoicec [seeed-2mic-voicecard]"

pi@raspberrypi:~ $ arecord -f cd -Dhw:2 | aplay -Dhw:2

Configure sound settings and adjust volume with alsamixer:

• alsamixer is a graphical mixer program for the Advanced Linux Sound Architecture (ALSA) that is used to configure sound settings and adjust the volume.

pi@raspberrypi:~ $ alsamixer

• For adjusting sound on ReSpeaker 2-Mics Pi HAT press F6 then select sound card. Select seeed-2mic-voicecard. Now you can configure sound and adjust volume as shown below.

sudo pip install spidev
cd ~/
git clone https://github.com/respeaker/mic_hat.git
cd mic_hat
python pixels.py

pi@raspberrypi:~ $ sudo pip3 install apa102-pi

from apa102_pi.driver import apa102
import time

# Initialize the library and the strip
strip = apa102.APA102(num_led=3, global_brightness=20, mosi=10, sclk=11,
                      order='rbg')

# Turn off all pixels (sometimes a few light up when the strip gets power)
strip.clear_strip()

# Prepare a few individual pixels
# strip.set_pixel_rgb(LED #, Color Value)
strip.set_pixel_rgb(0, 0xFF0000)  # Red
strip.set_pixel_rgb(1, 0xFFFFFF)  # White
strip.set_pixel_rgb(2, 0x00FF00)  # Green

# Copy the buffer to the Strip (i.e. show the prepared pixels)
strip.show()

# Wait a few Seconds, to check the result
time.sleep(20)

# Clear the strip
strip.clear_strip()
strip.cleanup()

• Run it and you can see the APA102 LEDs change color as red, white, and green.

Use on board User Button (Connected to GPIO17):

• Install driver by using this command on terminal.

sudo pip install rpi.gpio

• Test this code user_button.py or copy below code.

import RPi.GPIO as GPIO
import time

# Print Start of test
print("ReSpeaker 2-Mic Pi Button test")

# User button pin
BUTTON = 17
# User button setup
GPIO.setmode(GPIO.BCM)
GPIO.setup(BUTTON, GPIO.IN)

# Save previous state
previousState = GPIO.input(BUTTON)

# Unending Loop (main)
while True:
    # Get button state
    currentState = GPIO.input(BUTTON)
    # Check if any difference
    if currentState != previousState:
        # Store current state as previous
        previousState = currentState
        # Check the current state of the button
        if currentState:
            # Print if button is not clicked
            print("Button is not clicked")
        else:
            # Print if clicked
            print("Button is clicked")

• Run it and it should display on the terminal like these.

ReSpeaker 2-Mic Pi Button test
Button is clicked
Button is not clicked
Button is clicked
Button is not clicked
Button is clicked
Button is not clicked
Button is clicked
Button is not clicked

Voice Recording using ReSpeaker 2 Mic Pi Hat:

• Install driver by using this command on terminal.

sudo pip install pyaudio

• Test this code record_audio.py or copy code below.

import pyaudio
import wave
import numpy as np
from subprocess import call

class RecordAudio:
    '''
        Audio Recording
    '''
    # Initialize PyAudio
    pyaud = pyaudio.PyAudio()

    
    def record(self, rate, channels, width, deviceIndex,
            chunk, recordSeconds, fileName, extractChannel):
        '''
            Record Audio
        '''
        
        # Record settings
        stream = self.pyaud.open(
                    rate=rate,
                    format= self.pyaud.get_format_from_width(width),
                    channels=channels,
                    input=True,
                    input_device_index=deviceIndex)

        # Print start of recording
        print("Recording audio")
        
        # Get audio data and save as frame buffer.
        frames = []
        for i in range(0, int(rate / chunk * recordSeconds)):
            # Audio data
            data = stream.read(chunk)
            # For extracted channel
            if extractChannel is not None:
                # Get data for extracted channel
                a = np.frombuffer(data,dtype=np.int16)[extractChannel::channels]
                frames.append(a.tostring())
            # For using both channels
            else:
                # Get audio data from channels
                frames.append(data)
        
        # Close recording
        stream.stop_stream()
        stream.close()
        self.pyaud.terminate()

        # Print end of recording
        print("Done recording, now saving file...")
        
        # Saving file
        wf = wave.open(fileName, 'wb')
        if extractChannel is not None:
            wf.setnchannels(1)
        else:
            wf.setnchannels(channels)
        wf.setsampwidth(self.pyaud.get_sample_size(self.pyaud.get_format_from_width(width)))
        wf.setframerate(rate)
        wf.writeframes(b''.join(frames))
        wf.close()

        print("Done recording and file being saved.")
    
    def getAudioDeviceInfo(self):
        '''
            Audio Device Info
        '''
        # Settings to get device audio info
        info = self.pyaud.get_host_api_info_by_index(0)
        numdevices = info.get('deviceCount')
 
        for i in range(0, numdevices):
                if (self.pyaud.get_device_info_by_host_api_device_index(0, i).get('maxInputChannels')) > 0:
                    print("Input Device id ", i, " - ", self.pyaud.get_device_info_by_host_api_device_index(0, i).get('name'))

    def playAudio(self, fileName):
        '''
            Play Audio through aplay
        '''
        call(["aplay", fileName])

# Print Start of test
print("ReSpeaker 2-Mic Pi Audio test")

# Test of audio recording
recordAudio = RecordAudio()

# Audio Settings
RESPEAKER_RATE = 16000
RESPEAKER_CHANNELS = 2 
RESPEAKER_WIDTH = 2
CHUNK = 1024
RECORD_SECONDS = 5
WAVE_OUTPUT_FILENAME = "recorded_audio.wav"
# Refer to input device ID by running recordAudio.getAudioDeviceInfo()
RESPEAKER_INDEX = 0
# Extract data for specific channel. 
# Channel 1 set it to 0
# channel 2 set it to 1
# If no extracted channel set it to None
EXTRACT_CHANNEL = None

# Run get audio device infos.
# recordAudio.getAudioDeviceInfo()

# Run record audio
recordAudio.record(RESPEAKER_RATE,RESPEAKER_CHANNELS,RESPEAKER_WIDTH, 
RESPEAKER_INDEX, CHUNK, RECORD_SECONDS, 
WAVE_OUTPUT_FILENAME, EXTRACT_CHANNEL)

print("Playing recorded audio..")

# Play recorded audio
recordAudio.playAudio(WAVE_OUTPUT_FILENAME)

• Run the python code, check if you encounter this error:

OSError: [Errno -9998] Invalid number of channels

• If you encounter above error, change the value of RESPEAKER_INDEX = 0 to your own device ID. Sample shown below:

Input Device id  0  -  seeed-2mic-voicecard: bcm2835-i2s-wm8960-hifi wm8960-hifi-0 (hw:2,0)

• Run code again. Audio recording is successful when you see similar terminal output below, and your recorded audio is played.

ReSpeaker 2-Mic Pi Audio test
Input Device id  0  -  seeed-2mic-voicecard: bcm2835-i2s-wm8960-hifi wm8960-hifi-0 (hw:2,0)
Input Device id  4  -  capture
Input Device id  6  -  array
Input Device id  8  -  default
Recording audio
Done recording, now saving file...
Done recording and file being saved.
Playing recorded audio..
Playing WAVE 'recorded_audio.wav' : Signed 16 bit Little Endian, Rate 16000 Hz, Stereo

• To extract data by channel, you can edit this part of the python code:

# Extract data for specific channel. 
# Channel 1 set it to 0
# channel 2 set it to 1
# If no extracted channel set it to None
EXTRACT_CHANNEL = None

Libraries Used

https://github.com/respeaker/seeed-voicecard.git

https://github.com/respeaker/mic_hat.git

https://github.com/tinue/apa102-pi

https://github.com/createlabz/createlabz-public/blob/main/raspberry/respeaker/2-mic-pi-hat/

Conclusion

This device is useful in multiple ways. It can be as a voice recorder, voice recognition for security, as Alexa or Google Assistance, and many more. There are multiple ideas that this device can be used. 

References

https://wiki.seeedstudio.com/ReSpeaker_2_Mics_Pi_HAT/