Medv3: Synthetic Turkish Medical Speech Corpus for ASR Research

A synthetic speech corpus for Turkish medical automatic speech recognition research. Clinical sentences were synthesized using Google Cloud Text-to-Speech Chirp 3 HD voices. The dataset was created by turkmedstt and was last updated on June 10, 2026.

Use Cases

• Training Turkish medical speech recognition models based on synthetic clinical speech.
• Benchmarking ASR system performance on medical terminology mentioned in the description.
• Developing synthetic data pipelines for low-resource language medical ASR based on the described synthesis method.

Strengths

• All recordings are synthetic, avoiding privacy concerns associated with real patient data.
• Uses a specific, high-quality synthesis engine (Google Cloud Text-to-Speech Chirp 3 HD).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is synthetic, which may not capture all nuances of real-world clinical speech.

Provenance

Source

huggingface

Collection Method

Synthesized using Google Cloud Text-to-Speech Chirp 3 HD.

Freshness

Last updated 2026-06-10 16:21:30; freshness should be verified.