SpeechEditBench - Content: leaderboard
SpeechEditBench (2026) tests speech language models on editing a recording from a natural-language instruction in English and Mandarin, scored by automatic checks and an audio judge. This board covers replacing, inserting or deleting words. The score is joint success in percent: the edit must succeed and the spoken words must be preserved; higher is better.
Metric: Joint success (%): the requested edit is applied (task-specific automatic check) and the linguistic content is preserved (Whisper large-v3 or Paraformer transcript within 10% WER or CER of the expected text); source speech plus a natural-language instruction, English and Mandarin samples, each model through its released editing or conversation interface; content editing (replace, insert or delete a word or phrase; the target span must appear in or vanish from the ASR transcript, so joint success equals edit success), 600 samples; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3 Omni 30B A3B Instruct | 72 | #904 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=speecheditbench-content · How It Works · Data refreshed daily, snapshot 2026-11-13.