CohereLabs / cohere-transcribe-03-2026

Cohere Transcribe

Cohere Transcribe is an open source release of a 2B parameter dedicated audio-in, text-out, automatic speech recognition (ASR) model. The model supports 14 languages.

Developed by: Cohere and Cohere Labs .

• Point of Contact: Cohere Labs
• License: Apache 2.0 without any addendums
• Model: CohereLabs/cohere-transcribe-03-2026
• Model Size: 2B

Try Cohere Transcribe

You can try out the model in our Hugging Face Space .

Model Details

Input: Audio waveform. Audio is automatically resampled to 16kHz if necessary during preprocessing. Similarly, multi-channel (stereo) inputs are averaged to produce a single channel signal.

Output: Text.

Model Architecture: Cohere Transcribe is built on a speech-optimized Transformer variant: a Conformer. Input audio waveforms are converted into a mel-spectrogram and then processed by a Conformer encoder that holds the majority of the model’s parameters. The encoder’s representations are then passed to a lightweight Transformer decoder that generates text tokens. Cohere Transcribe is trained using standard supervised cross-entropy.

Languages covered: The model supports 14 languages: English, German, French, Italian, Spanish, Portuguese, Greek, Dutch, Polish, Arabic, Vietnamese, Chinese, Japanese and Korean.

Usage

CohereLabs/cohere-transcribe-03-2026 exposes a single model.transcribe() method that automatically handles long-form audio chunking and exposes parameters to facilitate efficient inference. It is recommended that you let the transcribe method handle batching for you. This implementation is optimized for offline inference: for online inference, see the vLLM integration example below.

Installation

Recommended (best offline throughput):

pip install "transformers>=4.56,<5.3,!=5.0.*,!=5.1.*" torch huggingface_hub soundfile librosa sentencepiece protobuf
pip install datasets  # only needed for examples 2 and 3

pip install "transformers>=4.52,!=5.0.*,!=5.1.*" torch huggingface_hub soundfile librosa sentencepiece protobuf

Example 1: Quick Start

Transcribe any audio file in a few lines. The model accepts file paths directly — no manual preprocessing required.

import torch
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
from huggingface_hub import hf_hub_download

model_id = "CohereLabs/cohere-transcribe-03-2026"

device = "cuda:0" if torch.cuda.is_available() else "cpu"

processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, trust_remote_code=True).to(device)
model.eval()

audio_file = hf_hub_download(
    repo_id="CohereLabs/cohere-transcribe-03-2026",
    filename="demo/voxpopuli_test_en_demo.wav",
)

texts = model.transcribe(processor=processor, audio_files=[audio_file], language="en")
print(texts[0])

import torch
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
from huggingface_hub import hf_hub_download


model_id = "CohereLabs/cohere-transcribe-03-2026"

device = "cuda:0" if torch.cuda.is_available() else "cpu"

processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, trust_remote_code=True).to(device)
model.eval()

from datasets import load_dataset

ds = load_dataset("google/fleurs", "ja_jp", split="test", streaming=True)
ds_iter = iter(ds)
samples = [next(ds_iter) for _ in range(3)]  # take 3 samples

audio_arrays = [s["audio"]["array"] for s in samples]
sample_rates = [s["audio"]["sampling_rate"] for s in samples]

# compile=True incurs a one-time warmup cost on the first call; subsequent calls are faster.
texts = model.transcribe(
    processor=processor,
    audio_arrays=audio_arrays,
    sample_rates=sample_rates,
    language="ja",
    compile=True,
    pipeline_detokenization=True,
    batch_size=16,
)
for ref, hyp in zip([s["transcription"] for s in samples], texts):
    print(f"REF: {ref}\nHYP: {hyp}\n")

import torch
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
from huggingface_hub import hf_hub_download


model_id = "CohereLabs/cohere-transcribe-03-2026"

device = "cuda:0" if torch.cuda.is_available() else "cpu"

processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, trust_remote_code=True).to(device)
model.eval()

from datasets import load_dataset

ds = load_dataset("distil-whisper/earnings22", "full", split="test", streaming=True)
sample = next(iter(ds))

import time

audio_array = sample["audio"]["array"]
sr = sample["audio"]["sampling_rate"]
duration_s = len(audio_array) / sr
print(f"Audio duration: {duration_s / 60:.1f} minutes")

start = time.time()
texts = model.transcribe(
    processor=processor,
    audio_arrays=[audio_array],
    sample_rates=[sr],
    language="en",
    compile=True,
)
elapsed = time.time() - start
rtfx = duration_s / elapsed
print(f"Transcribed in {elapsed:.1f}s — RTFx: {rtfx:.1f}")
print(f"Transcription ({len(texts[0].split())} words):")
print(texts[0][:500] + "...")

vLLM Integration

For production serving we recommend running via vLLM following the instructions below.

uv pip install -U vllm --torch-backend=auto --extra-index-url https://wheels.vllm.ai/nightly
uv pip install vllm[audio]
uv pip install librosa

vllm serve CohereLabs/cohere-transcribe-03-2026 --trust-remote-code

curl -v -X POST http://localhost:8000/v1/audio/transcriptions \
 -H "Authorization: Bearer $VLLM_API_KEY" \
-F "file=@$(realpath ${AUDIO_PATH})" \
-F "model=CohereLabs/cohere-transcribe-03-2026"

Strengths and Limitations

Cohere Transcribe is a performant, dedicated ASR model intended for efficient speech transcription.

Strengths

Cohere Transcribe demonstrates best-in-class transcription accuracy in 14 languages. As a dedicated speech recognition model, it is also efficient, benefitting from a real-time factor up to three times faster than that of other, dedicated ASR models in the same size range. The model was trained from scratch, and from the outset, we deliberately focused on maximizing transcription accuracy while keeping production readiness top-of-mind.

Limitations

• Single language. The model performs best when remaining in-distribution of a single, pre-specified language amongst the 14 in the range it supports. 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 Transcribe is eager to transcribe, even non-speech sounds. The model thus 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 [email protected] or raise an issue.

Terms of Use: We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant 2 billion parameter model to researchers all over the world. This model is governed by an Apache 2.0 license.