Instruction-Based Text-to-Speech Evaluation Benchmark

InstructTTSEval is a benchmark for evaluating Text-to-Speech systems on complex natural-language instructions. The dataset provides a hierarchical framework with three progressively challenging tasks testing acoustic control and style generalization. It was created by CaasiHUANG and last updated in June 2025.

Use Cases

• Benchmark TTS model performance on low-level acoustic control tasks using provided evaluation metrics.
• Evaluate high-level style generalization capabilities against the dataset's hierarchical instruction framework.
• Compare system outputs on the three progressively challenging tasks defined in the benchmark structure.

Strengths

• Hierarchical framework with three distinct evaluation tasks.
• Designed to test both acoustic control and style generalization.

Limitations

• Specific dataset size, row count, and audio sample details are not provided.
• The scope is limited to evaluation, not training, with no raw training data described.

Provenance

Source

CaasiHUANG via Hugging Face.

Collection Method

Created as an evaluation benchmark; specific data collection method not detailed.

Time Range

null

Freshness

Last updated on 2025-06 23.

Geography

null