Tasks and inputs
Cohere Transcribe Arabic is listed for Automatic speech recognition. It accepts audio and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Oct 7, 2026
Cohere Transcribe Arabic is a specialist speech-recognition model for Arabic dialects and Arabic–English code-switching.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
Cohere Transcribe Arabic is a specialist speech-recognition model for Arabic dialects and Arabic–English code-switching.
The publisher describes a 2B model focused on transcribing Arabic speech, including dialect variation and mixed Arabic–English speech. Its task is to return text from audio. This profile does not transfer speech-generation capabilities or general text-model context limits to the transcription checkpoint.
Build a trial set covering the dialects, microphones and speaking situations your product serves. Include spontaneous speech, a formal passage, names, numbers and a speaker who switches language within a sentence. Compare the transcript with a human reference and inspect substitutions that could change meaning, especially names, amounts and negation.
Follow the publisher’s audio preprocessing and checkpoint instructions. Long recordings may need segmentation; preserve timing and enough overlap to avoid losing words at boundaries. Distinguish transcription from speaker diarization and translation unless a chosen pipeline explicitly provides those additional tasks. For subtitles, review punctuation, line breaks and timing separately from word accuracy.
Keep the exact checkpoint identifier in the deployment configuration even when it contains a version suffix. Download access and licence terms belong to the linked checkpoint, while hosted availability and billing depend on the selected service. Do not present a per-minute or token rate without a documented host. Evaluate your own recordings before choosing a transcription service from results obtained on a different dialect mix.
Compare the model on the work you need to complete.
Cohere Transcribe Arabic is listed for Automatic speech recognition. It accepts audio and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.
Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.
The weights are downloadable under Apache 2.0. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
Gemini 3.5 Transcribe Live from Google: Speech transcription model for accurate audio-to-text and captioning workflows.
xAI
Grok Voice STT 1.0 from xAI: Grok Voice STT 1.0 is xAI's speech-to-text model. It supports transcription with word-level timestamps, optional speaker diarization, and multichannel audio.
OpenAI
GPT Realtime Whisper from OpenAI: Streaming speech-to-text model for low-latency transcript deltas from live audio.
openai
openai/whisper-large-v3-turbo is a source-linked Hugging Face repository assigned to Automatic speech recognition. Its recorded task interface accepts audio and produces text.
OpenAI
Whisper 3 Large from OpenAI: Open Whisper checkpoint for robust multilingual transcription and captioning.
OpenAI
Whisper Large v3 Turbo from OpenAI: Speech transcription model for accurate audio-to-text and captioning workflows.