Which one Performs Better? Wav2Vec or Whisper? Applying both in Badini Kurdish Speech to Text (BKSTT)
This work addresses a gap in speech-to-text technology for the Badini Kurdish dialect, an under-resourced language with about two million speakers, enabling better access to digital tools and visibility, though it is incremental as it applies existing methods to new data.
The researchers tackled the lack of speech-to-text systems for the Badini Kurdish dialect by creating and evaluating language models using Wav2Vec2-Large-XLSR-53 and Whisper-small on a dataset of kids' stories, finding that Wav2Vec2 achieved 90.38% readability and 82.67% accuracy, outperforming Whisper-small at 65.45% and 53.17%.
Speech-to-text (STT) systems have a wide range of applications. They are available in many languages, albeit at different quality levels. Although Kurdish is considered a less-resourced language from a processing perspective, SST is available for some of the Kurdish dialects, for instance, Sorani (Central Kurdish). However, that is not applied to other Kurdish dialects, Badini and Hawrami, for example. This research is an attempt to address this gap. Bandin, approximately, has two million speakers, and STT systems can help their community use mobile and computer-based technologies while giving their dialect more global visibility. We aim to create a language model based on Badini's speech and evaluate its performance. To cover a conversational aspect, have a proper confidence level of grammatical accuracy, and ready transcriptions, we chose Badini kids' stories, eight books including 78 stories, as the textual input. Six narrators narrated the books, which resulted in approximately 17 hours of recording. We cleaned, segmented, and tokenized the input. The preprocessing produced nearly 15 hours of speech, including 19193 segments and 25221 words. We used Wav2Vec2-Large-XLSR-53 and Whisper-small to develop the language models. The experiments indicate that the transcriptions process based on the Wav2Vec2-Large-XLSR-53 model provides a significantly more accurate and readable output than the Whisper-small model, with 90.38% and 65.45% readability, and 82.67% and 53.17% accuracy, respectively.