This systematic review examines the use of audio analysis and artificial intelligence (AI) for diagnosing respiratory diseases. The review focuses on three main areas: cough detection, lower respiratory symptom identification, and diagnostics from voice and speech. It analyzes 75 relevant studies, highlighting the trends, methods, and datasets used in each area. The review also discusses the impact of the COVID-19 pandemic on research, particularly in cough and respiratory sound classification. Key findings include the effectiveness of various machine learning (ML) algorithms, such as deep neural networks and statistical models, in identifying respiratory diseases. The review emphasizes the importance of data acquisition methods, feature extraction techniques, and the choice of ML models in achieving accurate diagnoses. Additionally, it highlights the potential of AI in providing quick and accurate diagnoses, reducing the need for invasive procedures, and enabling early treatment and continuous patient monitoring. The review concludes by discussing the limitations and future directions of research in this field.This systematic review examines the use of audio analysis and artificial intelligence (AI) for diagnosing respiratory diseases. The review focuses on three main areas: cough detection, lower respiratory symptom identification, and diagnostics from voice and speech. It analyzes 75 relevant studies, highlighting the trends, methods, and datasets used in each area. The review also discusses the impact of the COVID-19 pandemic on research, particularly in cough and respiratory sound classification. Key findings include the effectiveness of various machine learning (ML) algorithms, such as deep neural networks and statistical models, in identifying respiratory diseases. The review emphasizes the importance of data acquisition methods, feature extraction techniques, and the choice of ML models in achieving accurate diagnoses. Additionally, it highlights the potential of AI in providing quick and accurate diagnoses, reducing the need for invasive procedures, and enabling early treatment and continuous patient monitoring. The review concludes by discussing the limitations and future directions of research in this field.