M K Junayed P.
Before 365:
Product development | Bikroy
After 365:
Assistant Manager (Analyst) | Green Delta Insurance PLC
See all reviews
Online Course
Speech Recognition with Python
with
Ivan Manov
Master speech recognition—the technology that enables machines to understand human speech by converting voice into readable data. Utilize Python speech recognition tools to transcribe audio to text with cutting-edge AI models.
728 students already enrolled
Skill level:
Intermediate
Duration:
3 hours
-
Lessons (3 hours)
-
Practice exams (25 minutes)
- Lessons (3 hours)
- Practice exams (25 minutes)
CPE credits:
7
CPE stands for Continuing Professional Education and represents the mandatory credits a wide range of
professionals must earn to maintain their licenses and stay current with regulations and best
practices. One CPE credit typically equals 50 minutes of learning. For more details, visit NASBA's
official website: www.nasbaregistry.org
CPE stands for Continuing Professional Education and represents the mandatory credits a wide range of
professionals must earn to maintain their licenses and stay current with regulations and best
practices. One CPE credit typically equals 50 minutes of learning. For more details, visit NASBA's
official website: www.nasbaregistry.org
Accredited
certificate
What you learn
- Master audio and signal processing for speech-to-text.
- Understand how machines (and humans) process and interpret speech.
- Convert unstructured audio data into text.
- Use deep learning and APIs for speech recognition in Python.
- Implement AI-powered text-to-speech in Jupyter Notebook.
Topics & tools
Artificial IntelligenceDeep LearningPythonSignal ProcessingTransformersSpectrogramsSound and Speech FundamentalsHidden Markov ModelsSpeech-to-TextText-to-SpeechNeural NetworksSound EngineeringWhisper AIAudio For Machine LearningTheory
Your instructor
AI & Data Science Course Creator | Sound Engineer
Track record
Bringing real-world expertise from leading global companies
Academic background
Master's degree, Sound Engineering
Media and recognition
- Author of Speech Recognition with Python on 365DS & Udemy
- Speaker at Power Platform Bootcamp 2025 on Audio AI
- Sound Engineer for TV/Film incl. The Bachelor & award-winners
Course OVERVIEW
Description
CPE Credits: 7
Field of Study: Information Technology
Delivery Method: QAS Self Study
Delivery Method: QAS Self Study
Our Speech Recognition with Python course explores the technology that powers modern voice-activated systems and AI tools like virtual assistants, automated transcription devices, and home devices. We break down the theory behind speech recognition, covering Python audio processing and machine learning aspects in an easy-to-understand format. Along the way, we demonstrate the use of the librosa library, showing you how to perform essential audio processing tasks that are key to preparing sound data for analysis. You’ll gain hands-on experience as you implement speech-to-text tools using cutting-edge AI models like OpenAI’s Whisper and Google’s Web Speech API. Additionally, you'll explore the appropriate use of popular speech recognition toolkits like Assembly AI, Meta's Wav2Letter, Mozilla DeepSpeech, and cloud-based solutions, such as Amazon Transcribe and Azure Speech, considering accessibility and costs.
This speech recognition course unravels the behind-the-scenes processes that drive speech recognition. We explain how various methodologies operate—from audio feature extraction and noise cleaning to deep learning and transformers. We also cover essential audio concepts, including sound wave properties, analog-to-digital conversion, acoustics fundamentals, and aspects of human hearing.
By the end of the course, you'll be fully equipped with the skills to examine the speech recognition technology in greater depth and understand the fundamentals needed to build your own AI-powered model.
This course—tailored for data analysts, scientists, audio engineers, AI enthusiasts, and anyone with a curious mind—demonstrates how to convert sound files into structured, text-based outputs for analysis.
Whether you’re working with audio data or exploring AI, the Speech Recognition with Python course equips you with the knowledge to effectively transform audio into actionable insights.
Prerequisites
- Python (version 3.8 or later), SpeechRecognition and PyAudio libraries, and a code editor or IDE (e.g., Jupyter Notebook, Spyder, or VS Code)
- Basic understanding of Python programming is required.
- No prior experience with audio processing or speech recognition is necessary.
Advanced preparation
Curriculum
41 lessons
61 exercises
2 exams
- 2. Sound and Speech Basics12 minExplore the essential sound and speech concepts needed to master speech recognition. Learn how humans hear and process audio and how this compares to machine audio interpretation. Uncover the mechanics of sound waves and how they transfer energy through a medium, with intriguing Physics concepts explained in a clear and accessible way.12 minExplore the essential sound and speech concepts needed to master speech recognition. Learn how humans hear and process audio and how this compares to machine audio interpretation. Uncover the mechanics of sound waves and how they transfer energy through a medium, with intriguing Physics concepts explained in a clear and accessible way.How Do Humans Recognize Speech? FreeExercise FreeFundamentals of Sound and Sound Waves FreeProperties of Sound WavesExercise
- 3. Analog to Digital Conversion10 minLearn about converting analog signals into digital ones—a necessary step in working with audio data. Explore critical concepts like sample rate, bit depth, and bit rate and understand their connection to the sampling and quantization of a signal. Gain insights into what AI or machine learning engineers need to know about audio signal processing and discover the steps involved in preparing audio data for AI applications and modeling.10 minLearn about converting analog signals into digital ones—a necessary step in working with audio data. Explore critical concepts like sample rate, bit depth, and bit rate and understand their connection to the sampling and quantization of a signal. Gain insights into what AI or machine learning engineers need to know about audio signal processing and discover the steps involved in preparing audio data for AI applications and modeling.Key Concepts: Sample Rate, Bit Depth, and Bit RateAudio Signal Processing for Machine Learning and AIExercise
- 4. Audio Feature Extraction for AI Applications22 minExamine how audio features are extracted from a rough audio signal. Discover how sound characteristics and properties can combine and form features for building speech recognition models. Learn the role of the Fourier transform and its impact on this technology.22 minExamine how audio features are extracted from a rough audio signal. Discover how sound characteristics and properties can combine and form features for building speech recognition models. Learn the role of the Fourier transform and its impact on this technology.Time-Domain Audio FeaturesFrequency-Domain and Time-Frequency-Domain Audio FeaturesExerciseTime-Domain Feature Extraction: Framing and Feature ComputationFrequency-Domain Feature Extraction: Fourier TransformExercise
- 5. Speech Recognition Mechanics39 minUnderstand how speech recognition systems operate—from identifying patterns in sound to predicting words and sentences. Explore the statistical and deep learning methods behind the speech-to-text pipeline, including CNNs, RNNs, LSTMs, and Transformers. Learn the steps to build a speech recognition model and how to select the right tool for your tasks.39 minUnderstand how speech recognition systems operate—from identifying patterns in sound to predicting words and sentences. Explore the statistical and deep learning methods behind the speech-to-text pipeline, including CNNs, RNNs, LSTMs, and Transformers. Learn the steps to build a speech recognition model and how to select the right tool for your tasks.Acoustic and Language ModelingHidden Markov Models (HMMs) and Traditional Neural NetworksExerciseDeep Learning Models: CNNs, RNNs, and LSTMsAdvanced Speech Recognition Systems: TransformersExerciseBuilding a Speech Recognition Model Part IBuilding a Speech Recognition Model Part IISelecting the Appropriate Speech Recognition ToolExerciseExpanding Beyond the Tools We've Covered
- 6. Setting Up the Environment17 minSet up your environment for audio machine learning and hands-on speech-to-text implementation. Learn to install Anaconda and Jupyter in a dedicated environment using the correct Python version. Equip your system with essential libraries and tools for audio processing and AI, including Librosa, OpenAI's Whisper, PyTorch, and more.17 minSet up your environment for audio machine learning and hands-on speech-to-text implementation. Learn to install Anaconda and Jupyter in a dedicated environment using the correct Python version. Equip your system with essential libraries and tools for audio processing and AI, including Librosa, OpenAI's Whisper, PyTorch, and more.Installing AnacondaSetting up a New EnvironmentInstalling Packages for Speech RecognitionImporting the Relevant Packages in Jupyter Notebook
- 7. Transcribing Audio with Google Web Speech API33 minMaster how to distinguish between different audio files and how to load, play, and visualize them in Jupyter Notebook. Dive into Python’s speech_recognition library and discover how it enables access to Google’s API for implementing speech-to-text functionality. Find out how to assess the accuracy of your models or transcriptions by calculating WER (Word Error Rate) and CER (Character Error Rate).33 minMaster how to distinguish between different audio files and how to load, play, and visualize them in Jupyter Notebook. Dive into Python’s speech_recognition library and discover how it enables access to Google’s API for implementing speech-to-text functionality. Find out how to assess the accuracy of your models or transcriptions by calculating WER (Word Error Rate) and CER (Character Error Rate).Audio File Formats for Speech RecognitionImporting Audio Files in Jupyter NotebookExerciseThe SpeechRecognition Library: Google Web Speech APIEvaluation Metrics: WER and CERCalculating WER and CER in PythonExercise
- 8. Background Noise and Spectrograms20 minBackground noise often disrupts speech recognition systems. Learn to manage and visualize it by extracting spectrograms emphasizing crucial frequencies in audio files.20 minBackground noise often disrupts speech recognition systems. Learn to manage and visualize it by extracting spectrograms emphasizing crucial frequencies in audio files.Understanding Noise in Audio FilesCreating a Spectrogram with PythonExerciseDealing with Background NoiseExercise
- 9. Transcribing Audio with OpenAI's Whisper24 minDive deep into the speech recognition technology by using OpenAI's Transformer-based Whisper model. Learn to transcribe multiple audio files simultaneously and efficiently store results in a table for comprehensive analysis.24 minDive deep into the speech recognition technology by using OpenAI's Transformer-based Whisper model. Learn to transcribe multiple audio files simultaneously and efficiently store results in a table for comprehensive analysis.Whisper AI: Transformer-Based Speech-to-TextHomework AssignmentTranscribing Multiple Audio Files from a DirectoryExerciseSaving Audio Transcriptions to CSV for Easy AnalysisReversing the Process: AI-Powered Text-to-SpeechExercisePractice exam
- 10. Final Discussion and Future Directions12 minSummarize our discussion on AI-driven speech recognition, including emerging trends and modern practices. Explore the potential ethical and technological challenges these systems encounter. Experience the future of speech recognition with exciting insights into innovative topics like edge computing, real-time translation and even how a supermassive black hole can be explored through sonification.12 minSummarize our discussion on AI-driven speech recognition, including emerging trends and modern practices. Explore the potential ethical and technological challenges these systems encounter. Experience the future of speech recognition with exciting insights into innovative topics like edge computing, real-time translation and even how a supermassive black hole can be explored through sonification.Modern Practices and ApplicationsChallenges and LimitationsThe Future of Speech Recognition with AI
- 11. Course exam60 min60 minCourse exam
Free lessons
1.1 Welcome to the World of Speech Recognition
5 min
1.2 Course Approach
4 min
1.3 How It All Started: Formants, Harmonics, and Phonemes
3 min
1.5 Development and Evolution
4 min
2.1 How Do Humans Recognize Speech?
3 min
2.3 Fundamentals of Sound and Sound Waves
3 min
94%
of AI and data science graduates
successfully change
or advance their careers.
$29,000
average salary increase
after moving to an AI and data science career
9 in 10
of our graduates landed a new AI & data job
after enrollment
ACCREDITED certificates
Craft a resume and LinkedIn profile you’re proud of—featuring certificates recognized by leading global
institutions.
Earn CPE-accredited credentials that showcase your dedication, growth, and essential skills—the qualities
employers value most.
Certificates are included with the Self-study learning plan.
How it WORKS
Student REVIEWS
