Turkish Podcast Audio and Transcript Collection

Podcast recordings and associated text in the Turkish language. The dataset was created by Aynursusuz and was last updated in March 2026. The specific volume of episodes and total runtime are unknown.

Use Cases

• Train automatic speech recognition models on Turkish audio from the podcast recordings.
• Fine-tune language models for Turkish using the podcast text transcripts.
• Analyze spoken Turkish language features like vocabulary and syntax from the text data.
• Develop audio-text alignment models for Turkish using synchronized podcast audio and transcripts.

Strengths

• Dataset is optimized for efficient access with Parquet file format.
• Includes both audio and text modalities for Turkish language content.

Limitations

• Unknown total number of podcast episodes or hours of audio limits assessment of scale.
• Potential bias towards the specific podcast's topics, speakers, and recording quality.

Provenance

Source

Hugging Face, uploaded by user Aynursusuz.

Collection Method

Collection from a Turkish podcast source; specific gathering method is unknown.

Time Range

null

Freshness

Last updated in March 2026.

Geography

Primarily Turkey, based on language.