aiaimodel.cards
← All benchmarks

Common Voice

We haven't written a description for this benchmark yet; see the card notes below for how each maker ran it.

Your workloadRAG over docs · 12K in · 600 out · 50K requests/mo · 30% cached
  1. 1GPT-4o mini TTS · OpenAI
    0.139 WER

“Best value”: no other model here reports a higher score at a lower cost for your workload. Scores are self-reported by each maker with different settings, so treat close calls as ties.

How each maker reported it

  • GPT-4o mini TTS blog, Text-to-speech charts

    Word error rate of the generated speech (lower is better), read from OpenAI's charts. The second column is the bar labelled gpt-4o-mini-tts in the Dec 2025 post, the snapshot the 2025-12-15 release replaced.

Only each model's own maker's number is shown; scores a maker reports for rival models are left out.