Standard Moroccan Amazigh Audio and Text with <1,000 Records

Standard Moroccan Amazigh audio recordings and text transcripts totaling fewer than 1,000 records, created by abdelhaqueidali and updated in March 2026. The dataset provides raw, unprocessed speech data for the development of Automatic Speech Recognition and Text-to-Speech models.

Use Cases

• Training Automatic Speech Recognition (ASR) models using the audio recordings and text transcripts
• Developing Text-to-Speech (TTS) systems by mapping text transcripts to audio files
• Phonetic analysis of Standard Moroccan Amazigh using the raw audio recordings

Strengths

• Licensed under CC BY 4.0 for open use
• Targets the specific Standard Moroccan Amazigh (zgh) language variant
• Includes both audio and text modalities for multimodal speech tasks

Limitations

• Small sample size of fewer than 1,000 records
• Audio files are unprocessed and raw, potentially containing background noise
• Lack of detailed speaker metadata or demographic information

Provenance

Source

abdelhaqueidali

Collection Method

Original recordings created by the author

Freshness

Updated March 2026.

Geography

Morocco