Granite 4.0 1B Speech: Compact, Multilingual, and Built for the Edge

Source: Hugging Face Blog

Overview

We’re excited to share Granite 4.0 1B Speech, the latest addition to IBM’s Granite Speech collection. Designed for enterprise applications on resource‑constrained devices, Granite 4.0 1B Speech is a compact speech‑language model built for multilingual automatic speech recognition (ASR) and bidirectional speech translation (AST).

Key highlights

• Size: Half the parameters of its predecessor, granite‑speech‑3.3‑2b.
• Languages: English, French, German, Spanish, Portuguese, and Japanese (new Japanese ASR support).
• Features: Keyword‑list biasing for better recognition of names and acronyms.
• Performance: Higher English transcription accuracy and faster inference via speculative decoding.
• Recognition: Ranked #1 on the OpenASR leaderboard.

Performance

Despite its small size, Granite 4.0 1B Speech achieves highly competitive results on standard English ASR benchmarks. Performance is measured using Word Error Rate (WER)—the percentage of words transcribed incorrectly—where lower scores indicate better accuracy.

Benchmark Results

Chart 1: Granite 4.0 1B Speech delivers competitively low WER across many benchmarks while being a small model.

Licensing and Usage

• License: Apache 2.0.
• Framework support: Native integration with Transformers and vLLM.
• Evaluation: The model has been evaluated across a range of standard ASR and AST benchmarks—spanning English, multilingual, and translation tasks—and performs as well as or better than larger models.

For full evaluation results, architecture details, training data, and usage examples, see the model card.

Production Recommendations

We recommend pairing Granite 4.0 1B Speech with Granite Guardian for deployments that require additional risk detection.

Give it a try today and let us know what you think!