Implement chunk training

[Anatole-DC]

We are currently translating the audio to read all data one after the other, while keeping the dimensionality of the columns.

Tasks

• Preprocess the data into chunk
• Adapt training algorithm
• 🧪 Test if the training work on the new dataset