What is Speech-to-text?
Speech-to-text is a multimodal concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.
Definition
Speech-to-text is a multimodal concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system. The term is used in multimodal discussions, where precision matters: two teams can use the same word to mean slightly different things. In this glossary, speech-to-text is treated as a practical concept rather than a formal specification.
Why it matters
Multimodal terms describe how models handle images, audio, video and other non-text inputs. They matter because real applications increasingly mix media, and text-only benchmarks do not capture that behaviour.
Example
Suppose you are comparing Kimi K2 Chat with another model for a production task. Speech-to-text is one of the factors behind the numbers you see: Kimi K2 Chat lists a 400,000-token context window, costs $5.74 per million input tokens and scores 79.6 on MMMU. Understanding the term helps you decide which of those differences actually matter for your workload.
How to evaluate it
- Check whether speech-to-text is defined the same way in the documentation you are reading.
- Ask which measurement or trade-off the term is describing before comparing numbers.
- Separate vendor claims from independently observed behaviour on your own data.
- Write down the definition your team uses so reviews and evaluations stay consistent.
Related terms
Related guides
FAQ
What does Speech-to-text mean?
Speech-to-text is a multimodal concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.
Why does Speech-to-text matter?
Multimodal terms describe how models handle images, audio, video and other non-text inputs. They matter because real applications increasingly mix media, and text-only benchmarks do not capture that behaviour.
How is Speech-to-text used in practice?
Suppose you are comparing Kimi K2 Chat with another model for a production task. Speech-to-text is one of the factors behind the numbers you see: Kimi K2 Chat lists a 400,000-token context window, costs $5.74 per million input tokens and scores 79.6 on MMMU. Understanding the term helps you decide which of those differences actually matter for your workload.