Vietnamese Inverse Text Normalization Data for ASR

This dataset supports Vietnamese Inverse Text Normalization (ITN), a task that transforms spoken-style text to written form, particularly for improving Automatic Speech Recognition (ASR) output readability. It was created by VietAI and last updated in July 2022.

Use Cases

• Train sequence-to-sequence models to convert spoken-style Vietnamese text to its written form.
• Benchmark ITN model performance on Vietnamese language data.
• Improve ASR post-processing pipelines by integrating ITN for proper name correction.

Strengths

• Created by VietAI, an organization focused on Vietnamese AI.
• Last updated on July 12, 2022, providing a specific snapshot.

Limitations

• Specific data volume, column structure, and sample data are unavailable for assessment.
• The dataset's temporal coverage and geographic scope are not specified.
• The license and exact file formats are unknown.

Provenance

Source

VietAI

Freshness

Last updated 2022-07-12.

Geography

Vietnam (inferred from language focus)