MOSS-TTSD v1.0: Fused Text-to-Speech and Audio Tokenizer Model

OpenMOSS fused the MOSS-TTSD and MOSS-Audio-Tokenizer models for use with the SGLang framework. The dataset likely contains audio data and corresponding text or tokenized representations for speech synthesis tasks. Specific details on size, format, and creation date are not provided in the available metadata.

Use Cases

• Train text-to-speech models based on the fused MOSS-TTSD architecture mentioned in the description.
• Experiment with audio tokenization for generation tasks based on the included MOSS-Audio-Tokenizer.
• Benchmark speech synthesis performance using the SGLang framework referenced in the description.
• Fine-tune pre-trained audio models on custom voice data using the provided model components.

Strengths

• Integrates two specialized models (TTSD and Audio-Tokenizer) for a combined pipeline.
• Designed for compatibility with the SGLang framework, which may offer performance optimizations.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and file size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

OpenMOSS

Collection Method

Likely model fusion of existing components.