Is recording videos for AI legit in Nigeria?

Yes, recording videos for AI training data is a legitimate way to earn money in Nigeria. Companies like TrustAm work with AI developers who need real-world data. However, it's important to work with reputable platforms and understand that payment is for approved minutes, not guaranteed hours. Always be wary of any platform that asks for payment to join or for "activation fees."

What kind of work can I record for AI data?

You can record many types of everyday work and conversations. This includes trades like tailoring, auto mechanics, phone repair, cooking, hair braiding, welding, carpentry, and market trading. Speech data can include conversations in various Nigerian languages like English, Pidgin, Yoruba, Igbo, and Hausa. The key is to record your actual work or daily interactions.

How do I ensure my phone is suitable for recording AI data?

Your phone needs a good back camera capable of recording 1080p video and sufficient storage space (around 8GB free is recommended). Android 9 or newer is generally required. A strong battery is also helpful, though recordings can be set to upload over Wi-Fi to save mobile data. You can find more details on phone requirements on our blog.

Why is good lighting so important for AI recordings?

AI systems learn from visual patterns. Poor lighting makes it difficult for the AI to distinguish details, colours, and shapes. Inconsistent or flickering light can confuse the algorithms, leading to errors in recognition. Even lighting ensures that the AI receives clear, consistent visual input, which is essential for accurate learning and performance.

Can I record conversations with friends or family?

Yes, you can record conversations, but it's crucial that all participants give their consent before the recording begins. This ensures that you are complying with data privacy regulations. TrustAm requires explicit consent from all speakers involved in conversation recordings.

What happens if my recording is rejected?

If a minute of your recording is rejected, it means it did not meet the quality standards required by AI developers. This could be due to poor lighting, shaky video, unclear audio, or other technical issues. You are only paid for approved minutes. Understanding the reasons for rejection, as detailed in our review process articles, can help you improve your future recordings.

Good Light, Steady Camera: 7 Habits That Get AI Recordings Approved

Are you recording your work or conversations for AI training data and wondering why some of your minutes don't get approved? TrustAm pays for approved recordings, and understanding what makes a recording acceptable is key. This guide covers seven essential habits that help ensure your video and audio submissions meet the quality standards AI companies need. Following these tips will help you get more of your recorded work approved and paid for. This article is mainly for people who want to earn with their phone by recording their everyday activities.

Overview & Key Takeaways

Learn the seven key habits that ensure your AI training data recordings are approved and paid for. This includes maintaining good lighting, steady camera work, clear audio, and adhering to specific recording guidelines. Understanding these practices helps maximise earnings and contributes high-quality data for AI development.

Key Takeaways

Always ensure good, consistent lighting.
Keep the camera steady, preferably using a head strap.
Prioritise clear audio with minimal background noise.
Follow instructions for showing the random code and hands at the start.
Avoid covering the camera lens or phone during recording.
Record in batches, focusing on quality over quantity.
Understand the review process and why some minutes are cut.

Why Does Recording Quality Matter for AI Training Data?

AI systems, especially those using computer vision and speech recognition, learn from the data they are trained on. If the data is blurry, noisy, or inconsistent, the AI will not learn effectively. For example, an AI designed to understand car mechanics needs clear footage of tools being used and parts being fixed. Poor lighting can obscure crucial details, while shaky video makes it impossible to follow actions. Similarly, an AI learning Nigerian languages needs clear speech, free from interference. TrustAm collects this data to help AI developers build better, more reliable systems. Understanding what makes data useful is the first step to getting your recordings approved. This is why following guidelines is so important for anyone looking to earn from AI data in Nigeria.

Nigerian auto mechanic hands working on an engine, first-person view — Clear, steady footage from a mechanic's perspective is vital for AI training.

Habit 1: Master Your Lighting

Good lighting is perhaps the most critical factor for video recording approval. AI models need to see clearly, and this means avoiding both too little and too much light. Inconsistent lighting, like flickering from NEPA light or shadows moving across the scene, can also cause problems. Aim for natural, even light whenever possible. If you're recording indoors, position yourself near a window during the day. If you're working at night or in a poorly lit area, consider using a simple, consistent light source like a ring light or a well-placed lamp. Avoid recording directly in front of a bright light source, as this can wash out your face and the details of your work.

Habit 2: Keep the Camera Steady

A shaky camera is one of the quickest ways to get your recordings rejected. AI systems that process visual data rely on stable input to recognise patterns and actions. The best way to achieve this is by using a head-mounted phone holder. This keeps the camera fixed to your perspective as you work, providing a smooth, first-person view. If you don't have a head strap, try to mount your phone on a stable surface or hold it as still as possible. Avoid jerky movements or sudden pans. Remember, the goal is to mimic how a robot might perceive the world, and robots need steady input.

Nigerian tailor using a head-mounted phone for AI recording — A head strap ensures a steady, first-person view for AI data collection.

Habit 3: Prioritise Clear Audio

For speech data, clear audio is as important as clear video. Background noise, echoes, or muffled speech can make recordings unusable. Try to record in a quiet environment, away from traffic, loud conversations, or machinery. If you're recording conversations, ensure all participants are close to the phone's microphone. Speak clearly and at a moderate pace. For tasks that involve reading a random code at the start, enunciate each digit distinctly. If you're recording your work, minimise any sounds that might drown out your voice or the sounds of your trade, unless those sounds are specifically requested. Good audio quality is essential for training AI models to understand diverse Nigerian languages and accents accurately. This is why attention to detail in recording, as outlined in how reviews work, is so crucial.

Habit 4: Follow the Start-Up Protocol

Every recording session begins with specific instructions designed to verify the recording and the recorder. This typically involves stating a random three-digit code and showing both hands. It’s vital to perform this step correctly and clearly. Ensure the digits of the code are spoken distinctly and that your hands are fully visible. The random code acts as a unique identifier for the recording session, and showing your hands helps verify the recording's authenticity. Missing or fumbling this initial step can lead to the entire recording being flagged or rejected.

Habit 5: Keep the Lens Clear and Unobstructed

It sounds simple, but accidentally covering the camera lens with your finger, thumb, or even a stray hair is a common reason for rejection. Pay close attention to how you hold your phone or wear your head strap to ensure the camera has a clear, unobstructed view at all times. If you're recording yourself performing a task, regularly check that your body or equipment isn't blocking the lens. A clean lens also matters – wipe it with a soft cloth if it’s smudged or dirty before you start recording. This ensures that the AI receives a clear visual input, vital for tasks like recording clear hand movements.

Habit 6: Record in Batches, Focus on Quality

While it might be tempting to record for hours continuously, it's often more effective to record in focused batches. This allows you to concentrate on maintaining good lighting, a steady camera, and clear audio for each session. It also makes it easier to manage your phone's storage and battery life, and to upload recordings efficiently, as discussed in uploading AI recordings without finishing your mobile data. After each batch, you can quickly review your footage to catch any obvious errors before submitting. This approach helps ensure that the data you submit is of high quality, increasing the likelihood of approval and payment.

Habit 7: Understand the Review Process

TrustAm has a trained review team that checks each approved minute of recording. They look for issues like poor lighting, shaky video, background noise, and whether the recorded content matches the task requirements. Understanding that not every minute might be approved is part of the process. Minutes that don't meet the quality standards are cut, and you are paid only for the approved minutes. This rigorous review process ensures that the data supplied to AI companies is of the highest possible quality. Familiarising yourself with how the review process works can help you avoid common mistakes and improve your recording habits over time.