Technical Setup & Recording Environment
App Configuration
- Download Easy Voice Recorder (iOS/Android)
- Set format to WAV (PCM), 48 kHz sample rate, Mono channel. If the recordings don't have the right settings.
- Verify recordings save to internal storage or SD card
- One recording per sentence, stop and start for each sentence
- You can only do the project once.
Optimal Recording Environment
- Choose a quiet indoor room with minimal echo
- Avoid background noise, pets, traffic sounds
- Position phone microphone 30 cm from mouth
- Maintain fixed position throughout recording
Setting up Easy Voice Recorder
1:
3:
2:
4:
Recording Process & Pronunciation
Begin with Silence
Start each recording with 1-1.5 seconds of silence before speaking.
Clear Articulation
Read each sentence in your native Italian dialect at a conversational volume. Keep the pause between wake word and command under 0.3 seconds.
End with Silence
Allow 1-1.5 seconds of silence after speaking before stopping the recording.
Stop & Continue
Stop recording for each sentence. Start it again to proceed to the next numbered sentence. This makes sure its one sentence per file.
Listen to reference clips before recording to match intonation and pacing. Wake words include Kia, Hyundai, Genesis, Mercedes, and Gleo.
Google Docs
How to pronounce "Gleo"
Google Docs
How to pronounce "Genesis"
Google Docs
How to pronouce "Kia"
You are ready to start the recordings!
File Management & Quality Requirements
Naming Convention
Name files sequentially as 1.wav through 200.wav.
Android & iOS (Easy Voice Recorder)
- Open the “Listen" screen"
- Long-press one of the recordings, a bar on top appears → tap "Select all".
- Press the Google drive button → tab "Upload".
- Wait for 100 % upload, then close the app.
- Navigate to the personal folder you uploaded to.
- Move the files over to the folder we shared with you.
Quality Checklist
No coughing, popping sounds, or mouth clicks
Consistent microphone distance and volume
No filters, noise reduction, or compression
No background music, TV, or other speakers
One file per sentence
1:
3:
5:
2:
4:
