Fongbe Speech Dataset: A Tone-Preserved Continuous Speech Corpus

Professor's Fongbe Speech Dataset is a unified, high-quality collection of Fongbe speech data curated to preserve the linguistic integrity of this tonal language. It acts as a complete, unsegmented, and tone-accurate assembly of the Fongbe Continuous Speech Recognition corpora, merging the foundational ALFFA Project data from 2016 with an expanded Zenodo release from 2022. The dataset was last updated on the Hugging Face platform in February 2026.

Use Cases

• Train automatic speech recognition models based on the described continuous, tone-accurate Fongbe speech.
• Conduct linguistic research on tonal languages based on the dataset's stated preservation of linguistic integrity.
• Benchmark speech processing systems for low-resource languages based on the unified collection of Fongbe data.
• Develop language preservation tools based on the high-quality, curated Fongbe audio corpus.

Strengths

• Described as a unified, high-quality collection specifically curated for linguistic integrity.
• Merges two established sources: the ALFFA Project (2016) and a Zenodo release (2022).
• Explicitly designed to preserve tone accuracy for this tonal language.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.
• Data may reflect geographic or source bias inherent to the original ALFFA and Zenodo collections.

Provenance

Source

Merges data from the ALFFA Project (2016) and a Zenodo release (2022), hosted by 'Professor' on Hugging Face.

Collection Method

Curated assembly of existing Fongbe Continuous Speech Recognition corpora.

Freshness

Last updated 2026-02-15 00:16:05; freshness should be verified.