HarperValleyBank: A Domain-Specific Spoken Dialog Corpus
arXiv:2010.13929
Abstract
We introduce HarperValleyBank, a free, public domain spoken dialog corpus. The data simulate simple consumer banking interactions, containing about 23 hours of audio from 1,446 human-human conversations between 59 unique speakers. We selected intents and utterance templates to allow realistic variation while controlling overall task complexity and limiting vocabulary size to about 700 unique words. We provide audio data along with transcripts and annotations for speaker identity, caller intent, dialog actions, and emotional valence. The data size and domain specificity makes for quick transcription experiments with modern end-to-end neural approaches. Further, we provide baselines for representation learning, adapting recent work to embed waveforms for downstream prediction tasks. Our experiments show that tasks using our annotations are sensitive to both the model choice and corpus size.
arXiv comment5 pages content, 1 page of references
References in corpus (10)
- Adam: A Method for Stochastic Optimization
- A Simple Framework for Contrastive Learning of Visual Representations
- VoxCeleb: a large-scale speaker identification dataset
- Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition
- The Microsoft 2017 Conversational Speech Recognition System
- A Survey of Available Corpora for Building Data-Driven Dialogue Systems
- Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination
- Advancing the State of the Art in Open Domain Dialog Systems through the Alexa Prize
- On Mutual Information in Contrastive Learning for Visual Representations
- Who Needs Words? Lexicon-Free Speech Recognition