Speech-RATE: Synthetic Audio Recordings of Radiology Reports

Speech-RATE is a curated dataset derived from CT-RATE, providing synthetically generated audio recordings of radiology reports. The dataset is authored by lbuess and was last updated on 2026-04-09. It is designed to support research at the intersection of speech and medical imaging.

Use Cases

• Train speech synthesis models based on synthetically generated audio recordings of radiology reports.
• Develop multimodal AI systems based on the combination of audio and original CT-RATE medical imaging data.
• Research audio-text alignment in a clinical domain based on the paired radiology report content.

Strengths

• Dataset is derived from the established CT-RATE dataset, suggesting a structured medical source.
• Audio recordings are described as 'high-quality' and 'synthetically generated' in the description.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count, file formats, and column-level documentation are absent.
• The dataset is described as 'finalizing', which may indicate it is not yet fully released.

Provenance

Source

Derived from the CT-RATE dataset.

Collection Method

Curated and synthetically generated.

Time Range

null

Freshness

Last updated 2026-04 09 20:21:32; freshness should be verified.

Geography

null