Gametime: Lightweight Streaming Splits for Spoken Language Model Prototyping

The Gametime benchmark dataset is designed for prototyping in text-to-speech, automatic speech recognition, and spoken language models. It was created by the gametime-benchmark organization and last updated on October 19, 2025. The dataset provides lightweight, streaming-friendly splits for evaluating temporal dynamics in spoken language models.

Use Cases

• Prototyping text-to-speech models based on the dataset's streaming-friendly audio splits
• Evaluating automatic speech recognition systems based on the spoken language model benchmark
• Testing temporal dynamics in spoken language models based on the dataset's stated purpose
• Benchmarking model performance on spoken language tasks based on the dataset's design

Strengths

• Dataset is structured for streaming, which suggests efficient data loading
• Last updated on 2025-10-19, indicating recent maintenance
• Associated with a published paper for detailed methodology and context

Limitations

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

Provenance

Source

huggingface dataset by gametime-benchmark

Collection Method

Methodology described in the associated paper 'Game-Time: Evaluating Temporal Dynamics in Spoken Language Models'

Time Range

null

Freshness

Last updated 2025-10-19 07:23:13

Geography

null