Lesson 1: How Speech Becomes Text: Sound Waves and Spectrograms
Free, no sign-up needed.
Last updated
Learn how AI turns speech into text and text into natural speech. See how spectrograms and models like Whisper work, why prosody makes AI voices sound human, how real-time voice assistants work and why speed matters, why voice cloning needs only a short sample and why consent matters, and how to judge any voice tool. No code, no math.
You talk to your phone and your words appear on screen. A video gets captions on its own. An AI voice reads a story with pauses and feeling, and a voice assistant answers you out loud in about a second. This short course explains how all of that works, in plain language, with no code and no math. You will start with how speech becomes text: how a microphone turns your voice into numbers, what a spectrogram is, and how modern speech recognition models like Whisper read that picture of sound and write words one piece at a time. Then you will flip it around and see how text becomes natural speech, why older computer voices sounded robotic, and why prosody, the rhythm, stress, and pitch of speech, is what makes neural text-to-speech sound human. The second half looks at voice AI in real life. You will compare the classic three-step voice assistant pipeline with newer speech-to-speech models, and learn why latency matters so much in a spoken conversation. You will see why voice cloning can work from a short sample, what a clone copies well and what it misses, and why consent is the core rule. You will finish by learning why accents, noise, names, and numbers cause most errors, what word error rate means, and how to test any voice tool on your own audio before you trust it. The course is free, and you earn a free certificate when you pass the final exam.
2 modules • 5 lessons
Finish every lesson and pass the final exam to earn this free, shareable certificate.
Certificate of Completion
has successfully completed
5 lessons · Final exam passed
Michail Ouroumis
Founder, FreeAcademy.ai
Sample preview. Your name appears on the certificate when you complete the course. Learn more
Teaching a class or training a team? Assign this course for free
AI voice covers two main jobs. Speech recognition turns your voice into a picture of sound called a spectrogram, then a neural network predicts the words. Text-to-speech does the reverse: it plans the rhythm and pitch of speech, then turns that plan into a sound wave.
No. The course uses plain language, simple diagrams, and everyday examples. There is no code and no math.
AI Voice & Audio is hands-on and shows you how to make voiceovers, transcripts, dubs, and podcasts with real tools. This course explains how the technology works underneath, so you understand why tools behave the way they do.
No. It explains why cloning can work from a short sample, what a clone can and cannot copy, and why consent matters. It does not walk through any cloning tool.
About 40 minutes across five short lessons. Each lesson has a quiz, and there is a final exam with a free certificate when you pass.
The one AI idea that ties everything together. Learn what 'multimodal' means, how a single AI model reads text, images, and voice together, what you can do with it, and where it shows up in ChatGPT, Gemini, and Claude. A short, no-code, no-math course for complete beginners.
A short, no-code course on protecting yourself while you use AI. Learn what you should never paste into a chatbot, how to turn off data training in ChatGPT, Gemini, and Claude, how to spot voice-clone scam calls, deepfakes, and AI phishing, and how to keep your family safe. Practical safety habits for everyday AI users.
Turn text into natural narration, clone a voice with consent, transcribe and caption audio, dub into other languages, and build a full podcast or voiceover workflow using free AI tools.
Learn the idea behind AI search, recommendations, and "chat with your documents" tools. See how embeddings turn meaning into numbers, why close means related, how semantic search beats keyword search, where embeddings show up in everyday apps, why document search sometimes misses, and how to judge tools built on it. No code, no math.
Some AI models pause to "think" before they answer. Learn what reasoning models actually do, why thinking step by step helps, how models are trained to reason, when a reasoning model is worth the extra time and cost, how to prompt one well, and where their thinking still goes wrong. No code, no math.
Understand how tools like Midjourney, DALL-E, and Stable Diffusion actually generate an image, in plain language with no code and no math. Learn what a diffusion model is, how starting from noise leads to a finished picture, what latent space is, how your prompt steers the process, and why diffusion beat older methods like GANs.
