The dataset contains real recordings that define a simple spoken language understanding task. You can download it from here. The Fluent Speech Commands dataset contains 30,043 utterances from 97 speakers. It is recorded as 16 kHz single-channel .wav files each containing a single utterance used for controlling smart-home appliances or virtual assistant, for example, “put on the music” or “turn up the heat in the kitchen”. Each audio is labeled with three slots: action, object, and location. A slot takes on one of the multiple values: for instance, the “location” slot can take on the values “none”, “kitchen”, “bedroom”, or “washroom”. We refer to the combination of slot values as the intent of the utterance. For each intent, there are multiple possible wordings: for example, the intent {action: “activate”, object: “lights”, location: “none”} can be expressed as “turn on the lights”, “switch the lights on”, “lights on”, etc. The dataset has a total of 248 phrasing mapping to 31 unique intents.

Attention-based RNN sequence-to-sequence model for the Fluent Speech Commands dataset. This model checkpoint achieves 99.6% accuracy on the test set. The model uses an ASR model trained on LibriSpeech (speechbrain/asr-crdnn-rnnlm-librispeech) to extract features from the input audio, then maps these features to an intent and slot labels using a beam search.

You can try the model on the example_fsc.wav file included here as follows:

The system is trained with recordings sampled at 16kHz (single channel). The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling decode_file if needed. Make sure your input tensor is compliant with the expected sampling rate if you use encode_batch and decode_batch.

To perform inference on the GPU, add run_opts={"device":"cuda"} when calling the from_hparams method.

The model was trained with SpeechBrain (f1f421b3). To train it from scratch follows these steps:

cd speechbrain
pip install -r requirements.txt
pip install -e .

cd recipes/fluent-speech-commands
python train.py hparams/train.yaml --data_folder=your_data_folder

You can find our training results (models, logs, etc) here.

The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.

SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to be simple, extremely flexible, and user-friendly. Competitive or state-of-the-art performance is obtained in various domains.

Website: https://speechbrain.github.io/

GitHub: https://github.com/speechbrain/speechbrain