Imperial EEE MEng Final Year Project : Speech-to-Speech Summarisation
Files include :
- MATLAB portion to extract speech features. Utilises Voicebox for pitch estimation.
- Python notebook for execution of Mode 1, 2 and 3 algorithms
- Python notebook for data exploration. Uses How2 Dataset transcriptions and youtube-dl to download wav files of speech. Also align transcription to the audio file.
- Python notebook for hosting WaveNet generation API on Colab.