AppTek Call-Center Dialogues - Mexican English: leaderboard
AppTek Call-Center Dialogues (2026) is a long-form English speech recognition test set of spontaneous role-played call-center conversations in 14 accents, recorded for evaluation only. This board is the word error rate in percent on the Mexican-accented English calls, with Silero segmentation; lower is better.
Metric: Word error rate (%) on the Mexican-accented English (en_MX) calls of AppTek Call-Center Dialogues (128.6 hours of spontaneous role-played English call-center conversations, 14 accents), default inference settings, scored per session after Open ASR Leaderboard text normalization, audio segmented with the Silero VAD; lower is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3-ASR-1.7B | 9.3 | |
| 2 | Qwen3-ASR-0.6B | 10.3 | |
| 3 | canary-qwen-2.5B | 10.6 | |
| 4 | Phi-4 Multimodal Instruct | 10.9 | #4154 |
| 5 | parakeet-tdt-0.6B-v3 | 10.9 | |
| 6 | parakeet-tdt-0.6B-v2 | 10.9 | |
| 7 | canary-1B-v2 | 12.2 | |
| 8 | granite-speech-3.3-8B | 12.6 | |
| 9 | kyutai-stt-pytorch-stt-2.6B-en | 13.2 | |
| 10 | granite-speech-3.3-2B | 13.2 | |
| 11 | whisper-large-v2 | 14.3 | |
| 12 | Whisper Large-v3 | 18.4 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=apptek-call-center-dialogues-mexican-english · How It Works · Data refreshed daily, snapshot 2026-11-21.