Music Audio and Text Dataset

A dataset containing audio and text data, hosted on Hugging Face by author NjNBrl. It was last updated in March 2026. Specific content details such as genre, instruments, or recording sources are not provided.

Use Cases

• Train audio classification models using the audio features to identify music genres or instruments.
• Develop automatic speech recognition or lyric transcription systems using the paired audio and text data.
• Build text-to-music generation models by learning the relationship between the text descriptions and audio samples.
• Perform music similarity analysis by extracting and comparing embeddings from the audio files.

Strengths

• Last updated in March 2026, indicating recent maintenance.
• Contains both audio and text modalities for multimodal tasks.

Limitations

• Unknown total size, row count, and specific column structure.
• No information on audio quality, sample rate, or recording conditions.
• Potential for bias or limited scope due to unspecified data collection method.

Provenance

Source

Hugging Face, author NjNBrl.

Freshness

Last updated 2026-03-21.

Geography

Region tags suggest US focus.