AppTek Call-Center Dialogues - Welsh 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 Welsh-accented English calls, with Silero segmentation; lower is better.

Metric: Word error rate (%) on the Welsh-accented English (en_GB_WLS) 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

#ModelScoreOverall rank
1Qwen3-ASR-1.7B9.5
2Qwen3-ASR-0.6B10.3
3canary-qwen-2.5B10.4
4parakeet-tdt-0.6B-v310.7
5parakeet-tdt-0.6B-v210.7
6Phi-4 Multimodal Instruct11.2#4154
7kyutai-stt-pytorch-stt-2.6B-en11.8
8granite-speech-3.3-8B12.1
9canary-1B-v212.2
10granite-speech-3.3-2B13.2
11Whisper Large-v316.6
12whisper-large-v216.6

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-welsh-english · How It Works · Data refreshed daily, snapshot 2026-11-21.