Sesotho Speech Synthesis Audio Recordings

OpenSLR 32 provides high-quality, multi-speaker transcribed audio data for Sesotho, one of four South African languages. The dataset consists of wave files and corresponding transcriptions in a TSV file. It was uploaded to Hugging Face by voice-biomarkers in August 2024.

Use Cases

• Train a text-to-speech model using the wave files and TSV transcriptions to generate synthetic Sesotho speech.
• Fine-tune a multilingual automatic speech recognition system on the transcribed Sesotho audio data.
• Analyze speaker characteristics and prosody across multiple speakers using the multi-speaker wave files.
• Benchmark speech synthesis quality for South African languages using the high-quality audio recordings.

Strengths

• High-quality audio recordings suitable for speech synthesis.
• Multi-speaker data provides voice diversity for model training.
• Includes transcriptions for supervised learning tasks.

Limitations

• Specific row count, audio duration, and number of speakers are unknown.
• Covers only one of the four South African languages mentioned in the source.
• Potential for speaker demographic bias is unverified.

Provenance

Source

OpenSLR (Open Speech and Language Resources), specifically SLR32.

Collection Method

null

Time Range

null

Freshness

Last updated on the platform in August 2024.

Geography

South Africa