使用 Xiao ESP32S3 对 reSpeaker Flex 进行 HTTP 流式传输
介绍
本指南介绍如何使用 ESP32-S3（XIAO ESP32S3）从 reSpeaker Flex XVF3800 麦克风阵列采集音频，录制 5 秒 16 kHz 立体声 32 位 PCM 音频，并将数据存储在 PSRAM 中以处理大容量音频缓冲区。随后通过 Wi-Fi 将录制的音频上传到基于 Python 的 Flask HTTP 服务器，在服务器端接收并保存为标准 WAV 文件。该方案非常适合语音采集、远程音频日志记录，以及包括语音转文本（STT）和基于机器学习的音频分析在内的语音处理流水线等应用。
| reSpeaker Flex XVF3800 线性阵列搭配 XIAO ESP32S3 | reSpeaker Flex XVF3800 环形阵列搭配 XIAO ESP32S3 | |
|---|---|---|
 |  | |
Arduino 代码
note
你必须启用 PSRAM，否则 malloc() 会失败。
在上传之前更新以下字段：
// WiFi credentials
const char* ssid = "your SSID";
const char* password = "your password";
// HTTP server (your PC)
const char* serverUrl = "http://<flask-server>:8000/upload"; // please replace with your PC IP /flask server
完整 Arduino 代码
#include "WiFi.h"
#include "HTTPClient.h"
#include "AudioTools.h"
// WiFi credentials
const char* ssid = "your SSID";
const char* password = "your password";
// HTTP server (your PC)
const char* serverUrl = "http://<flask-server>:8000/upload"; // please replace with your PC IP /flask server
// Audio: 16kHz, stereo, 32-bit
AudioInfo info(16000, 2, 32);
I2SStream i2s_in;
I2SConfig i2s_config;
// 5 seconds of audio = 640,000 bytes
#define RECORDING_SECONDS 5
#define BYTES_PER_SECOND 128000
#define TOTAL_BYTES (RECORDING_SECONDS * BYTES_PER_SECOND)
// Buffer for recording
uint8_t* audioBuffer = nullptr;
void connectWiFi() {
Serial.printf("Connecting to WiFi: %s\n", ssid);
WiFi.begin(ssid, password);
int attempts = 0;
while (WiFi.status() != WL_CONNECTED && attempts < 20) {
delay(500);
Serial.print(".");
attempts++;
}
if (WiFi.status() == WL_CONNECTED) {
Serial.println("\nConnected!");
Serial.printf("IP Address: %s\n", WiFi.localIP().toString().c_str());
} else {
Serial.println("\nFailed to connect!");
}
}
void setupI2SInput() {
i2s_config = i2s_in.defaultConfig(RX_MODE);
i2s_config.copyFrom(info);
// XVF3800 pins
i2s_config.pin_bck = 8;
i2s_config.pin_ws = 7;
i2s_config.pin_data = 44;
i2s_config.pin_data_rx = 43;
i2s_config.is_master = false;
i2s_in.begin(i2s_config);
Serial.println("I2S input started.");
}
void setup() {
Serial.begin(115200);
while(!Serial);
AudioLogger::instance().begin(Serial, AudioLogger::Info);
// Allocate memory for audio buffer
audioBuffer = (uint8_t*)malloc(TOTAL_BYTES);
if (!audioBuffer) {
Serial.println("Failed to allocate memory!");
return;
}
connectWiFi();
setupI2SInput();
// Wait for I2S to stabilize
delay(500);
// Record audio
Serial.printf("Recording %d seconds of audio...\n", RECORDING_SECONDS);
size_t total_read = 0;
size_t bytes_read = 0;
unsigned long start_time = millis();
while (total_read < TOTAL_BYTES) {
bytes_read = i2s_in.readBytes(audioBuffer + total_read,
min(4096, (int)(TOTAL_BYTES - total_read)));
if (bytes_read > 0) {
total_read += bytes_read;
// Progress indicator
if (total_read % BYTES_PER_SECOND == 0) {
Serial.printf("Recorded %.1f seconds\n", total_read / (float)BYTES_PER_SECOND);
}
}
}
unsigned long record_time = millis() - start_time;
Serial.printf("Recording complete! %d bytes in %lu ms\n", total_read, record_time);
// Send via HTTP POST
if (WiFi.status() == WL_CONNECTED) {
HTTPClient http;
Serial.printf("Sending audio to %s\n", serverUrl);
http.begin(serverUrl);
http.addHeader("Content-Type", "application/octet-stream");
http.addHeader("X-Sample-Rate", String(info.sample_rate));
http.addHeader("X-Channels", String(info.channels));
http.addHeader("X-Bits-Per-Sample", String(info.bits_per_sample));
int httpResponseCode = http.POST(audioBuffer, total_read);
if (httpResponseCode > 0) {
Serial.printf("HTTP Response code: %d\n", httpResponseCode);
String response = http.getString();
Serial.println("Response: " + response);
} else {
Serial.printf("Error code: %d\n", httpResponseCode);
Serial.println("Error: " + http.errorToString(httpResponseCode));
}
http.end();
} else {
Serial.println("WiFi not connected!");
}
// Free memory
free(audioBuffer);
Serial.println("Done!");
}
void loop() {
// Nothing - runs once
}
预期输出
Python 服务器代码（Flask）
你需要在你的环境中安装并运行 flask pip install flask
from flask import Flask, request, jsonify
import wave
import datetime
app = Flask(__name__)
@app.route('/upload', methods=['POST'])
def upload_audio():
try:
# Get audio parameters from headers
sample_rate = int(request.headers.get('X-Sample-Rate', 16000))
channels = int(request.headers.get('X-Channels', 2))
bits_per_sample = int(request.headers.get('X-Bits-Per-Sample', 32))
# Get raw audio data
audio_data = request.data
print(f"Received {len(audio_data)} bytes")
print(f"Format: {sample_rate}Hz, {channels} channels, {bits_per_sample}-bit")
# Generate filename with timestamp
timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
filename = f"recording_{timestamp}.wav"
# Save as WAV file
with wave.open(filename, 'wb') as wav_file:
wav_file.setnchannels(channels)
wav_file.setsampwidth(bits_per_sample // 8) # Convert bits to bytes
wav_file.setframerate(sample_rate)
wav_file.writeframes(audio_data)
duration = len(audio_data) / (sample_rate * channels * (bits_per_sample // 8))
print(f"Saved to {filename} ({duration:.2f} seconds)")
return jsonify({
'status': 'success',
'filename': filename,
'bytes_received': len(audio_data),
'duration_seconds': duration
}), 200
except Exception as e:
print(f"Error: {str(e)}")
return jsonify({'status': 'error', 'message': str(e)}), 500
@app.route('/', methods=['GET'])
def index():
return "Audio Upload Server Running"
if __name__ == '__main__':
print("Starting HTTP server on port 8000...")
print("Waiting for audio uploads from ESP32...")
app.run(host='0.0.0.0', port=8000, debug=True)
预期输出
技术支持与产品讨论
感谢你选择我们的产品！我们将为你提供多种支持，确保你在使用我们产品的过程中尽可能顺利。我们提供多种沟通渠道，以满足不同的偏好和需求。