Ported from upstream commit
0a8193c,
pinned 2026-07-07.
Validated against the Transformers reference at transcribe.cpp commit
d89ecb7
on 2026-07-07.
Offline Arabic speech-to-text, including dialectal Arabic and
Arabic-English code-switching, with English as a secondary language. An
Arabic-focused adaptation of the Cohere Transcribe 03-2026 architecture:
a Conformer encoder with a Transformer encoder-decoder head
(cross-attention, tied token embedding). Takes a 16 kHz mono WAV and a
language flag (-l ar or -l en) and produces a transcript. Decoding
is autoregressive.
WER measured on the full FLEURS Arabic (ar_eg) test split (428
utterances) with greedy decoding and no external LM, scored with the
Whisper BasicTextNormalizer. BF16 reference baseline, measured with
native Transformers on the same manifest: 11.00%; the BF16 port scores
11.02%, and every quant falls inside the reference's 95% confidence
interval. FLEURS Arabic is Egyptian-dialect speech; upstream numbers
published on other Arabic test sets are not directly comparable.
Cohere Transcribe Arabic is an open source release of a 2B parameter dedicated audio-in, text-out automatic speech recognition (ASR) model.
The optimized for Arabic and English, with a focus on Arabic dialect performance and Arabic-English code-switching.
Based on the Cohere Transcribe architecture.
audio waveform → log-Mel spectrogram. Audio is automatically resampled to 16kHz if necessary during preprocessing. Similarly, multi-channel (stereo) inputs are averaged to produce a single channel signal.
Output
transcribed text
Model
a large Conformer encoder extracts acoustic representations, followed by a lightweight Transformer decoder for token generation
Cohere Transcribe Arabic is supported natively in transformers. This is the recommended way to use the model for
offline inference. For online inference, see the vLLM integration example below.
For audio longer than the feature extractor's max_audio_clip_s, the feature extractor automatically splits the waveform into chunks.
The processor reassembles the per-chunk transcriptions using the returned audio_chunk_index.
Cohere Transcribe Arabic demonstrates strong transcription accuracy for Arabic and English. As a dedicated speech recognition model, it benefits from efficient inference via the Conformer encoder-decoder architecture.
Limitations
Single language. The model performs best when remaining in-distribution of a single, pre-specified language. It does not feature explicit, automatic language detection and exhibits inconsistent performance on code-switched audio.
Timestamps/Speaker diarization. The model does not feature either of these.
Silence. Like most AED speech models, Cohere Transcibe Arabic is eager to transcribe, even non-speech sounds. The model benefits from prepending a noise gate or VAD (voice activity detection) model in order to prevent low-volume, floor noise from turning into hallucinations.
Model Card Contact
For errors or additional questions about details in this model card, contact labs@cohere.com or raise an issue.
Terms of Use:
We hope that the release of this model will make community-based research efforts into Arabic speech more accessible. This model is governed by an Apache 2.0 license.
