TOSD: Tamazight Open Speech Dataset for ASR and TTS

A parsed and formatted voice dataset containing recordings and text transcripts in Standard Moroccan Amazigh. The dataset is intended for training Automatic Speech Recognition and Text-to-Speech models and was published by the Tamazight-NLP organization. The dataset page was last updated on April 21, 2026.

Use Cases

• Train Automatic Speech Recognition (ASR) models based on the provided voice recordings and transcripts.
• Develop Text-to-Speech (TTS) systems based on the paired audio and text data.
• Conduct linguistic research on the Standard Moroccan Amazigh language based on the speech corpus.

Strengths

• Data is described as parsed, formatted, and ready-to-use, which suggests preprocessing effort.
• Contains paired audio recordings and text transcripts, a core requirement for supervised speech model training.

Limitations

• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The description references a raw repository with additional data not included here, indicating this may be a subset.

Provenance

Source

Tamazight-NLP organization on Hugging Face.

Collection Method

Likely collected and curated for speech technology applications; specific gathering method is unknown.

Freshness

Last updated 2026-04-21 01:15:29; freshness should be verified.

Geography

The language covered is Standard Moroccan Amazigh, suggesting a primary focus on Morocco.