CapTTS-SFT: Style-Captioned Text-to-Speech Training Data

OpenSound created this dataset for training CapTTS, EmoCapTTS, and AccCapTTS models, as described in the paper 'CapSpeech: Enabling Downstream Applications in Style-Captioned Text-to-Speech'. The dataset was last updated on July 28, 2025. It contains audio-text pairs sourced from multiple original datasets.

Use Cases

• Train text-to-speech models based on audio-text pairs.
• Develop emotion-captioned speech synthesis based on the described EmoCapTTS task.
• Build accent-captioned speech synthesis based on the described AccCapTTS task.
• Research style transfer in speech synthesis based on the 'CapSpeech' paper framework.

Strengths

• Created for a published academic paper, suggesting a research-grade purpose.
• Designed for three specific, advanced text-to-speech tasks (CapTTS, EmoCapTTS, AccCapTTS).
• Last updated on July 28, 2025, indicating recent maintenance.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The actual audio files are hosted separately, requiring additional steps for access.

Provenance

Source

OpenSound

Collection Method

Aggregated from multiple original datasets.

Freshness

Last updated 2025-07-28 02:34:38