Arabic
Emerging7papers using it
2023first seen
The 'Arabic' dataset/benchmark contains ASR hypotheses and reference transcriptions used to evaluate the performance of automatic speech recognition systems through fine-grained, Part-of-Speech-wise error analysis.
Papers using Arabic (7)
- Breaking the Script Barrier: Enabling Automatic Alignment for PoS-based ASR Error Analysis in Non-Latin ScriptsFlavors of Moonshine: Tiny Specialized ASR Models for Edge DevicesHENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and TranslationSAGE: Spliced-Audio Generated Data for Enhancing Foundational Models in Low-Resource Arabic-English Code-Switched Speech RecognitionExploring Speech Enhancement for Low-resource Speech SynthesisAdvocating Character Error Rate for Multilingual ASR EvaluationSTTATTS: Unified Speech-To-Text And Text-To-Speech Model