# Audio
## Python Packages
Examples of audio data science projects are sentiment prediction, speech-to-text prediction, and music generation.
The article [1] discusses three Audio Data Science Project Python packages.
### Magenta
Magenta is an open-source Python package built on top of TensorFlow to manipulate image and music data to train a machine learning model with the generative model as the output.
Magenta does not provide clear API references for learning, but they do provide a lot of research demo and collaborator notebooks.
### Librosa
Librosa is a Python package developed for music and audio analysis.
Librosa is specific on capturing the audio information to be transformed into a data block. However, the documentation and example are good to understand how to work with audio data science projects.
### pyAudioAnalysis
pyAudioAnalysis is a Python package for audio analysis tasks such as:
- Extract Audio Features
- Train machine learning model for audio segmentation
- Classification of unknown audio
- Emotion recognition with a Regression model
- Dimensional Reduction for audio data visualization
There is a lot that can be done with this package.
## Speech Recognition
[Speech Recognition in Python - The Complete Beginner’s Guide](https://towardsdatascience.com/speech-recognition-in-python-the-complete-beginners-guide-de1dd7f00726)
[Create Searchable Audio using NLTK and Speech Recognition](https://betterprogramming.pub/create-searchable-audio-using-python-78b5afc5122)
[Machine Learning is Fun Part 6: How to do Speech Recognition with Deep Learning](https://medium.com/@ageitgey/machine-learning-is-fun-part-6-how-to-do-speech-recognition-with-deep-learning-28293c162f7a)
[Speech Recognition — GMM, HMM](https://jonathan-hui.medium.com/speech-recognition-gmm-hmm-8bb5eff8b196)
[Speech Recognition — Feature Extraction MFCC and PLP](https://jonathan-hui.medium.com/speech-recognition-feature-extraction-mfcc-plp-5455f5a69dd9)
[Speech Recognition — Acoustic, Lexicon and Language Model](https://jonathan-hui.medium.com/speech-recognition-acoustic-lexicon-language-model-aacac0462639)
[The Kinetics Dataset: Train and Evaluate Video Classification Models using FitftyOne](https://www.instapaper.com/read/1498461024)
## AssemblyAI
[AssemblyAI](https://www.assemblyai.com/)
[How to Perform Sentiment Analysis Over Audio Files with Python](https://towardsdatascience.com/sentiment-analysis-assemblyai-python-a4686967e0fc)
[How to Summarize Audio and Video Files with Python](https://towardsdatascience.com/summarize-audio-video-files-assemblyai-c9126918870c)
[How To Moderate Audio Data Using Python and AssemblyAI?](https://www.instapaper.com/read/1484566419)
[How to Perform Speech-To-Text and Remove PII with AssemblyAI](https://www.instapaper.com/read/1480619727)
## References
[1]: [Top 3 Python Packages to learn Audio Data Science Project](https://towardsdatascience.com/top-3-python-packages-to-learn-audio-data-science-project-cbd11c100fe7)
[2]: [Get To Know Audio Feature Extraction in Python](https://towardsdatascience.com/get-to-know-audio-feature-extraction-in-python-a499fdaefe42)
[3]: [Audio Augmentations in TensorFlow](https://towardsdatascience.com/audio-augmentations-in-tensorflow-48483260b169)