Fluent Speech Commands is an open source audio dataset for spoken language understanding (SLU) experiments. Each utterance is labeled with "action", "object", and "location" values; for example, "turn the lights on in the kitchen" has the label {"action": "activate", "object": "lights", "location": "kitchen"}. A model must predict each of these values, and a prediction for an utterance is deemed to be correct only if all values are correct. The task is very simple, but the dataset is large and flexible to allow for many types of experiments: for instance, one can vary the number of speakers, or remove all instances of a particular sentence and test whether a model trained on the remaining sentences can generalize.
A Bilingual Dataset for Bangla and English Voice Commands Colloquial Bangla has adopted many English words due to colonial influence. In conversational Bangla, it is quite common to speak in a mixture of English and Bangla. This phenomenon, prevalent in conversational language is known as code-switching (CS). CS is defined as the continuous alternation between two languages in a single conversation. Thus, in Bangla natural language processing, it is often necessary to map a single base command to its many different variants - spoken in multiple mixtures of English and Bangla. In order to facilitate this, we have curated a dataset centered around common browser commands.