Berlin EmoDB

A database of emotional speech. Ten actors (5 female and 5 male) simulated the emotions,<br> producing 10 German utterances (5 short and 5 longer sentences) which could be used in everyday communication<br> and are interpretable in all applied emotions.<br> The recordings were taken in an anechoic chamber with high-quality recording equipment. In addition to the sound<br> electro-glottograms were recorded. The speech material comprises about 800 sentences (seven emotions * ten actors * ten sentences + some second versions).<br> The complete database was evaluated in a perception test regarding the recognisability of emotions and their<br> naturalness. Utterances recognised better than 80% and judged as natural by more than 60% of the listeners were included in the database, resulting in 535 samples. They are split into 4 speakers in test and 6 speakers in train. The database meta information is in audformat. The audio format is 16kHz wav format. Here's a sample script in Python how to get train and test dataframes: <em>import audformat</em> <em>!wget https://zenodo.org/record/7447302/files/emodb.zip</em> <em>!unzip emodb.zip</em> <em>db = audformat.Database.load('emodb')</em> <em>df_test = db['emotion.categories.test.gold_standard'].get()</em> <em>df_train = db['emotion.categories.train.gold_standard'].get()</em> <em>print(f'samples in train: {df_train.shape[0]}, and test: {df_test.shape[0]}')</em>