Spoken Term Detection Methods for Sparse Transcription in Very Low-resource Settings
This addresses transcription challenges for linguists documenting oral languages with minimal data, though it is incremental as it builds on existing methods.
The paper tackled the problem of spoken term detection in very low-resource settings where only a few minutes of transcribed speech are available, showing that a fine-tuned pretrained universal phone recognizer outperforms dynamic time warping and that using a graph structure to represent phoneme ambiguity improves recall while maintaining high precision.
We investigate the efficiency of two very different spoken term detection approaches for transcription when the available data is insufficient to train a robust ASR system. This work is grounded in very low-resource language documentation scenario where only few minutes of recording have been transcribed for a given language so far.Experiments on two oral languages show that a pretrained universal phone recognizer, fine-tuned with only a few minutes of target language speech, can be used for spoken term detection with a better overall performance than a dynamic time warping approach. In addition, we show that representing phoneme recognition ambiguity in a graph structure can further boost the recall while maintaining high precision in the low resource spoken term detection task.