🎙️ Speech-to-Text on ESP32 using Wit.ai

Convert spoken words to text in real-time using an ESP32 microcontroller, an I2S MEMS microphone, and the Wit.ai speech recognition API — with results displayed on an OLED screen and the Serial Monitor.

---

📖 Overview

This project demonstrates how to build a low-cost, Wi-Fi-enabled Speech-to-Text system on the ESP32. Audio is captured via an I2S digital microphone, streamed to the Wit.ai cloud API for transcription, and the resulting text is shown on a connected OLED display.

---

✨ Features

• 🎤 Real-time audio capture using an I2S MEMS microphone
• ☁️ Cloud-based speech recognition via Wit.ai
• 🖥️ Transcribed text displayed on an SSD1306 OLED screen
• 📟 Output also available via Serial Monitor (for debugging)
• 📶 Wi-Fi connectivity using the ESP32's built-in radio

---

🛠️ Hardware Requirements

Component	Description
ESP32 Dev Board	Main microcontroller (e.g., ESP32-WROOM-32)
I2S MEMS Microphone	e.g., INMP441
OLED Display	0.96" SSD1306 (128×64, I2C)
Jumper Wires	For connections
Breadboard	Optional, for prototyping

---

🔌 Wiring Diagram

I2S Microphone → ESP32

Mic Pin	ESP32 Pin
VDD	3.3V
GND	GND
WS (LRCK)	GPIO 15
SCK (BCLK)	GPIO 14
SD (Data)	GPIO 32
L/R	GND (Left channel)

OLED Display → ESP32 (I2C)

OLED Pin	ESP32 Pin
VCC	3.3V
GND	GND
SDA	GPIO 21
SCL	GPIO 22

⚠️ Pin numbers may vary depending on your specific ESP32 board. Adjust in the code as needed.

---

📦 Software Requirements

• Arduino IDE (v1.8+ or v2.x)
• ESP32 Board Support Package
• Required Libraries:
  • Adafruit SSD1306
  • Adafruit GFX Library
  • WiFiClientSecure (built-in with ESP32 core)
  • ArduinoJson (optional, for parsing Wit.ai response)

---

⚙️ Setup & Installation

1. Clone the Repository

git clone https://github.com/your-username/esp32-speech-to-text.git
cd esp32-speech-to-text

2. Install ESP32 Board in Arduino IDE

1. Go to File → Preferences
2. Add this URL to Additional Board Manager URLs:

   https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_index.json
   

3. Go to Tools → Board → Board Manager, search for esp32, and install.

3. Install Required Libraries

In Arduino IDE, go to Sketch → Include Library → Manage Libraries and install:

• Adafruit SSD1306
• Adafruit GFX Library
• ArduinoJson (if used)

4. Configure Your Credentials

Open the main .ino file and update the following:

const char* ssid     = "YOUR_WIFI_SSID";
const char* password = "YOUR_WIFI_PASSWORD";
const char* witai_token = "YOUR_WIT_AI_ACCESS_TOKEN";

5. Get a Wit.ai Access Token

1. Go to https://wit.ai and sign in with your Facebook/Meta account.
2. Create a new app and select your language.
3. Copy the Server Access Token from Settings.

6. Upload to ESP32

1. Select your board under Tools → Board → ESP32 Dev Module
2. Select the correct Port
3. Click Upload

---

🚀 How It Works

[User Speaks]
     ↓
[I2S Mic captures audio]
     ↓
[ESP32 buffers audio samples]
     ↓
[Audio sent to Wit.ai via HTTPS POST]
     ↓
[Wit.ai returns transcribed text (JSON)]
     ↓
[Text displayed on OLED + Serial Monitor]

---

📁 Project Structure

esp32-speech-to-text/
├── esp32_speech_to_text.ino   # Main Arduino sketch
├── wit_ai.h                   # Wit.ai API communication
├── i2s_mic.h                  # I2S microphone configuration
├── oled_display.h             # OLED display helpers
└── README.md

---

🧪 Serial Monitor Output

Open the Serial Monitor at 115200 baud to see debug logs:

Connecting to WiFi...
Connected! IP: 192.168.1.42
Recording audio...
Sending to Wit.ai...
Response: "turn on the light"

---

🐛 Troubleshooting

Issue	Possible Fix
No audio captured	Check I2S wiring; verify pin definitions in code
OLED not displaying	Confirm I2C address (usually 0x3C); check SDA/SCL pins
Wi-Fi not connecting	Double-check SSID and password in config
Wit.ai returns empty	Speak clearly; check your access token; verify audio format
Upload fails	Hold the BOOT button on ESP32 during upload

---

📄 License

This project is licensed under the MIT License.

---

🙌 Acknowledgements

• Wit.ai — Free speech recognition API by Meta
• Espressif — ESP32 platform
• Adafruit — OLED display libraries

---

🤝 Contributing

Pull requests are welcome! For major changes, please open an issue first to discuss what you'd like to change.

1. Fork the repo
2. Create your feature branch: git checkout -b feature/my-feature
3. Commit your changes: git commit -m 'Add some feature'
4. Push to the branch: git push origin feature/my-feature
5. Open a Pull Request