pith. sign in

arxiv: 2505.13404 · v2 · pith:UWKI6DURnew · submitted 2025-05-19 · 💻 cs.CL · eess.AS

Granary: Speech Recognition and Translation Dataset in 25 European Languages

Nithin Rao Koluguri , Monica Sekoyan , George Zelenfroynd , Sasha Meister , Shuoyang Ding , Sofia Kostandian , He Huang , Nikolay Karpov
show 7 more authors
Jagadeesh Balam Vitaly Lavrukhin Yifan Peng Sara Papi Marco Gaido Alessio Brutti Boris Ginsburg
This is my paper
classification 💻 cs.CL eess.AS
keywords datalanguagestranslationdatasetsgranarymodelspipelinespeech
0
0 comments X
read the original abstract

Multi-task and multilingual approaches benefit large models, yet speech processing for low-resource languages remains underexplored due to data scarcity. To address this, we present Granary, a large-scale collection of speech datasets for recognition and translation across 25 European languages. This is the first open-source effort at this scale for both transcription and translation. We enhance data quality using a pseudo-labeling pipeline with segmentation, two-pass inference, hallucination filtering, and punctuation restoration. We further generate translation pairs from pseudo-labeled transcriptions using EuroLLM, followed by a data filtration pipeline. Designed for efficiency, our pipeline processes vast amount of data within hours. We assess models trained on processed data by comparing their performance on previously curated datasets for both high- and low-resource languages. Our findings show that these models achieve similar performance using approx. 50% less data. Dataset will be made available at https://hf.co/datasets/nvidia/Granary

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. S-JEPA : Soft Clustering Anchors for Self-Supervised Speech Representation Learning

    cs.SD 2026-06 unverdicted novelty 7.0

    S-JEPA uses soft GMM posteriors in a JEPA framework for self-supervised speech learning, achieving lowest WER below 90M parameters without offline re-clustering.

  2. The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning

    eess.AS 2026-03 unverdicted novelty 7.0

    FLAIR enables spoken dialogue AI to conduct continuous latent reasoning while perceiving speech through recursive latent embeddings and an ELBO-based finetuning objective.

  3. TRADE: Transducer-Augmented Decoder for Speech LLM

    cs.CL 2026-06 unverdicted novelty 6.0

    TRADE augments multimodal Speech LLMs with a transducer branch for streaming ASR, reporting 6.71% WER offline and 8.40% streaming on the Open ASR Leaderboard from one checkpoint.

  4. AfriVox-v2: A Domain-Verticalized Benchmark for In-the-Wild African Speech Recognition

    cs.CL 2026-05 unverdicted novelty 6.0

    AfriVox-v2 is a benchmark that evaluates modern speech models on in-the-wild African audio with domain-specific tests for sectors including government, finance, health, and agriculture.

  5. Reducing the Offline-Streaming Gap for Unified ASR Transducer with Consistency Regularization

    eess.AS 2026-04 unverdicted novelty 6.0

    A single RNN-T model with chunked context and consistency regularization improves low-latency streaming ASR accuracy while keeping offline performance intact.

  6. Rethinking Entropy Allocation in LLM-based ASR: Understanding the Dynamics between Speech Encoders and LLMs

    eess.AS 2026-04 unverdicted novelty 6.0

    A multi-stage training method for LLM-based ASR uses new entropy allocation metrics to achieve competitive benchmark performance with 2.3B parameters while mitigating hallucinations via better encoder-LLM decoupling.

  7. OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models

    cs.CL 2026-04 unverdicted novelty 6.0

    OmniVoice introduces a diffusion language model-style non-autoregressive TTS system that directly maps text to multi-codebook acoustic tokens, scaling zero-shot synthesis to over 600 languages with SOTA results on mul...

  8. Rethinking Depth: A study of the Recursive-Transformer for Speech Recognition

    eess.AS 2026-06 unverdicted novelty 5.0

    Experimental study finds Recursive-Transformer for ASR encoders achieves comparable performance with 66% fewer parameters when limited recursion is applied in the latent space.

  9. Comparing Human and Automatic Recognition of Dutch Dysarthric Continuous Speech: A Case Study

    cs.CL 2026-06 unverdicted novelty 4.0

    Case study finds that fine-tuned ASR models outperform human listeners on Dutch dysarthric continuous speech from one speaker, lowering WER from over 70% to over 23%.

  10. Identification and Anonymization of Named Entities in Unstructured Information Sources for Use in Social Engineering Detection

    cs.LG 2026-04 unverdicted novelty 4.0

    A Telegram data collection pipeline with Parakeet transcription and custom transformer NER models that detects sensitive entities at high F1 scores and provides anonymization metrics preserving structural coherence.

  11. The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning

    eess.AS 2026-03 unverdicted novelty 4.0

    FLAIR enables simultaneous latent reasoning during speech input in full-duplex dialogue models via recursive latent embeddings and an ELBO-based training objective without added latency.