Lab3: Audio Recording and Playback

Introduction

In this lab, you will learn how to record and play audio on the Mini Pupper using Python. The Mini Pupper 2 has built-in audio capabilities that allow you to record sound through a microphone and play it back through speakers. This is useful for voice commands, audio feedback, and interactive applications.

Prerequisites

Install the required Python packages:

pip install sounddevice soundfile

Audio Configuration

Before running the audio test, ensure your audio devices are properly configured. Check or create the ~/.asoundrc file:

nano ~/.asoundrc

Add the following configuration:

pcm.!default {
    type asym
    playback.pcm {
        type plug
        slave.pcm "hw:0,0"  # Playback device
    }
    capture.pcm {
        type plug
        slave.pcm "hw:1,0"  # Capture device
    }
}

ctl.!default {
    type hw
    card 0  # Control device
}

This configuration sets up asymmetric audio with separate playback and capture devices.

Running the Audio Test

Navigate to the demos folder and run the test:

cd ~/mini_pupper_bsp/demos/
python audio_test.py

Example Output

Complete Audio Test Code

#!/usr/bin/env python3
import sounddevice as sd
import soundfile as sf
import time
import os

# Audio record parameters
fs = 48000  # 48KHz, Audio sampling rate
duration = 5  # Recording duration in seconds

# Set the default speaker volume to maximum
# Headphone number is 0 without HDMI output
# Headphone number is 1 when HDMI connect the display
os.system("amixer -c 0 sset 'Headphone' 100%")

print("Mini Pupper 2 audio record start...")
record = sd.rec(int(duration * fs), samplerate=fs, channels=2)
sd.wait()  # Wait for record to finish
print("Mini Pupper 2 audio record end.")

# Increase the volume [manually]
record *= 100

# Save the record
sf.write('/tmp/mini-pupper-2-audio_test.wav', record, fs)

# Wait for 1 second to playback
time.sleep(1)

# Play audio
print("Mini Pupper 2 audio playback start...")
sd.play(record[:, 0], fs)
sd.wait()  # Wait for playback to finish
sd.play(record[:, 1], fs)
sd.wait()  # Wait for playback to finish
print("Mini Pupper 2 audio playback end.")

Code Breakdown

1. Import Libraries

import sounddevice as sd
import soundfile as sf
import time
import os

sounddevice: Python library for audio recording and playback
soundfile: Library for reading and writing audio files
time: For adding delays
os: For system commands

2. Audio Parameters

fs = 48000  # 48KHz, Audio sampling rate
duration = 5  # Recording duration in seconds

fs: Sample rate of 48kHz (CD quality is 44.1kHz)
duration: How long to record in seconds

3. Set Speaker Volume

os.system("amixer -c 0 sset 'Headphone' 100%")

Uses ALSA mixer to set the headphone volume to maximum. The card number may vary:

Card 0: Without HDMI output
Card 1: When HDMI is connected to a display

4. Record Audio

record = sd.rec(int(duration * fs), samplerate=fs, channels=2)
sd.wait()

sd.rec(): Records audio with specified samples, sample rate, and channels
sd.wait(): Blocks until recording is complete
channels=2: Stereo recording

5. Amplify and Save

record *= 100
sf.write('/tmp/mini-pupper-2-audio_test.wav', record, fs)

Multiply the recording by 100 to increase volume
Save as WAV file to /tmp/ directory

6. Playback Audio

sd.play(record[:, 0], fs)
sd.wait()
sd.play(record[:, 1], fs)
sd.wait()

Play left channel (record[:, 0])
Play right channel (record[:, 1])
Wait for each playback to complete

Exercises

Exercise 1: Variable Recording Duration

Modify the script to accept recording duration as a command-line argument.

import sys
duration = int(sys.argv[1]) if len(sys.argv) > 1 else 5

Exercise 2: Audio Level Meter

Create a real-time audio level meter that displays the input volume.

import numpy as np

def audio_callback(indata, frames, time, status):
    volume = np.linalg.norm(indata) * 10
    print("|" + "=" * int(volume) + " " * (50 - int(volume)) + "|")

with sd.InputStream(callback=audio_callback):
    sd.sleep(10000)

Exercise 3: Voice-Activated Recording

Implement voice-activated recording that starts when sound is detected above a threshold.

Troubleshooting

Issue	Solution
No audio devices found	Check `arecord -l` and `aplay -l` for available devices
Permission denied	Add user to audio group: `sudo usermod -a -G audio $USER`
Wrong device	Update card numbers in `~/.asoundrc`
No sound output	Check volume with `alsamixer`
Recording too quiet	Increase the multiplication factor (e.g., `record *= 200`)

Checking Audio Devices

List recording devices:

arecord -l

List playback devices:

aplay -l

Test recording:

arecord -d 5 -f cd test.wav

Test playback:

aplay test.wav

Summary

In this lab, you learned:

How to configure ALSA audio devices on Raspberry Pi
How to record audio using the sounddevice library
How to save and load audio files with soundfile
How to play back recorded audio
How to adjust audio volume programmatically