Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2507.13264

Voxtral

Alexander H. Liu·Andy Ehrenberg·Andy Lo·Clément Denoix·Corentin Barreau·Guillaume Lample·Jean-Malo Delignon·Khyathi Raghavi Chandu·Patrick von Platen·Pavankumar Reddy Muddireddy·Sanchit Gandhi·Soham Ghosh·Srijan Mishra·Thomas Foubert·Abhinav Rastogi·Adam Yang·Albert Q. Jiang·Alexandre Sablayrolles·Amélie Héliou·Amélie Martin·Anmol Agarwal·Antoine Roux·Arthur Darcet·Arthur Mensch·Baptiste Bout·Baptiste Rozière·Baudouin De Monicault·Chris Bamford·Christian Wallenwein·Christophe Renaudin·Clémence Lanfranchi·Darius Dabert·Devendra Singh Chaplot·Devon Mizelle·Diego de las Casas·Elliot Chane-Sane·Emilien Fugier·Emma Bou Hanna·Gabrielle Berrada·Gauthier Delerce·Gauthier Guinet·Georgii Novikov·Guillaume Martin·Himanshu Jaju·Jan Ludziejewski·Jason Rute·Jean-Hadrien Chabran·Jessica Chudnovsky·Joachim Studnia·Joep Barmentlo·Jonas Amar·Josselin Somerville Roberts·Julien Denize·Karan Saxena·Karmesh Yadav·Kartik Khandelwal·Kush Jain·Lélio Renard Lavaud·Léonard Blier·Lingxiao Zhao·Louis Martin·Lucile Saulnier·Luyu Gao·Marie Pellat·Mathilde Guillaumin·Mathis Felardos·Matthieu Dinot·Maxime Darrin·Maximilian Augustin·Mickaël Seznec·Neha Gupta·Nikhil Raghuraman·Olivier Duchenne·Patricia Wang·Patryk Saffer·Paul Jacob·Paul Wambergue·Paula Kurylowicz·Philomène Chagniot·Pierre Stock·Pravesh Agrawal·Rémi Delacourt·Romain Sauvestre·Roman Soletskyi·Sagar Vaze·Sandeep Subramanian·Saurabh Garg·Shashwat Dalal·Siddharth Gandhi·Sumukh Aithal·Szymon Antoniak·Teven Le Scao·Thibault Schueller·Thibaut Lavril·Thomas Robert·Thomas Wang·Timothée Lacroix·Tom Bewley·Valeriia Nemychnikova·Victor Paltz
View original source ↗Read source document ↗Export metadata ↓

Abstract

We present Voxtral Mini and Voxtral Small, two multimodal audio chat models. Voxtral is trained to comprehend both spoken audio and text documents, achieving state-of-the-art performance across a diverse range of audio benchmarks, while preserving strong text capabilities. Voxtral Small outperforms a number of closed-source models, while being small enough to run locally. A 32K context window enables the model to handle audio files up to 40 minutes in duration and long multi-turn conversations. We also contribute three benchmarks for evaluating speech understanding models on knowledge and trivia. Both Voxtral models are released under Apache 2.0 license.

Explore related subjects

cs.SDcs.AIeess.AS

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alexander H. Liu, Andy Ehrenberg, Andy Lo, Clément Denoix, Corentin Barreau, Guillaume Lample, Jean-Malo Delignon, Khyathi Raghavi Chandu, Patrick von Platen, Pavankumar Reddy Muddireddy, Sanchit Gandhi, Soham Ghosh, Srijan Mishra, Thomas Foubert, Abhinav Rastogi, Adam Yang, Albert Q. Jiang, Alexandre Sablayrolles, Amélie Héliou, Amélie Martin, Anmol Agarwal, Antoine Roux, Arthur Darcet, Arthur Mensch, Baptiste Bout, Baptiste Rozière, Baudouin De Monicault, Chris Bamford, Christian Wallenwein, Christophe Renaudin, Clémence Lanfranchi, Darius Dabert, Devendra Singh Chaplot, Devon Mizelle, Diego de las Casas, Elliot Chane-Sane, Emilien Fugier, Emma Bou Hanna, Gabrielle Berrada, Gauthier Delerce, Gauthier Guinet, Georgii Novikov, Guillaume Martin, Himanshu Jaju, Jan Ludziejewski, Jason Rute, Jean-Hadrien Chabran, Jessica Chudnovsky, Joachim Studnia, Joep Barmentlo, Jonas Amar, Josselin Somerville Roberts, Julien Denize, Karan Saxena, Karmesh Yadav, Kartik Khandelwal, Kush Jain, Lélio Renard Lavaud, Léonard Blier, Lingxiao Zhao, Louis Martin, Lucile Saulnier, Luyu Gao, Marie Pellat, Mathilde Guillaumin, Mathis Felardos, Matthieu Dinot, Maxime Darrin, Maximilian Augustin, Mickaël Seznec, Neha Gupta, Nikhil Raghuraman, Olivier Duchenne, Patricia Wang, Patryk Saffer, Paul Jacob, Paul Wambergue, Paula Kurylowicz, Philomène Chagniot, Pierre Stock, Pravesh Agrawal, Rémi Delacourt, Romain Sauvestre, Roman Soletskyi, Sagar Vaze, Sandeep Subramanian, Saurabh Garg, Shashwat Dalal, Siddharth Gandhi, Sumukh Aithal, Szymon Antoniak, Teven Le Scao, Thibault Schueller, Thibaut Lavril, Thomas Robert, Thomas Wang, Timothée Lacroix, Tom Bewley, Valeriia Nemychnikova, Victor Paltz. 2025-07-17. Voxtral. https://arxiv.org/abs/2507.13264

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Streaming Generation for Music Accompaniment

Music generation models can produce high-fidelity coherent accompaniment given complete audio input, but are limited to editing and loop-based workflows. We study real-time audio-to-audio accompaniment: as a model hears an input audio stream (e.g., a singer singing), it has to also simultaneously generate in real-time a coherent accompanying stream (e.g., a guitar accompaniment). In this work, we propose a model design considering inevitable system delays in practical deployment with two design variables: future visibility $t_f$, the offset between the output playback time and the latest input time used for conditioning, and output chunk duration $k$, the number of frames emitted per call. We train Transformer decoders across a grid of $(t_f,k)$ and show two consistent trade-offs: increasing effective $t_f$ improves coherence by reducing the recency gap, but requires faster inference to stay within the latency budget; increasing $k$ improves throughput but results in degraded accompaniment due to a reduced update rate. Finally, we observe that naive maximum-likelihood streaming training is insufficient for coherent accompaniment where future context is not available, motivating advanced anticipatory and agentic objectives for live jamming.

cs.SD↗
arXiv2026

M-CIF: Multi-Scale Alignment For CIF-Based Non-Autoregressive ASR

The Continuous Integrate-and-Fire (CIF) mechanism provides effective alignment for non-autoregressive (NAR) speech recognition. This mechanism creates a smooth and monotonic mapping from acoustic features to target tokens, achieving performance on Mandarin competitive with other NAR approaches. However, without finer-grained guidance, its stability degrades in some languages such as English and French. In this paper, we propose Multi-scale CIF (M-CIF), which performs multi-level alignment by integrating character and phoneme level supervision progressively distilled into subword representations, thereby enhancing robust acoustic-text alignment. Experiments show that M-CIF reduces WER compared to the Paraformer baseline, especially on CommonVoice by 4.21% in German and 3.05% in French. To further investigate these gains, we define phonetic confusion errors (PE) and space-related segmentation errors (SE) as evaluation metrics. Analysis of these metrics across different M-CIF settings reveals that the phoneme and character layers are essential for enhancing progressive CIF alignment.

cs.SD↗
arXiv2026

Decoding Order Matters in Autoregressive Speech Synthesis

Autoregressive speech synthesis often adopts a left-to-right order, yet generation order is a modelling choice. We investigate decoding order through masked diffusion framework, which progressively unmasks positions and allows arbitrary decoding orders during training and inference. By interpolating between identity and random permutations, we show that randomness in decoding order affects speech quality. We further compare fixed strategies, such as \texttt{l2r} and \texttt{r2l} with adaptive ones, such as Top-$K$, finding that fixed-order decoding, including the dominating left-to-right approach, is suboptimal, while adaptive decoding yields better performance. Finally, since masked diffusion requires discrete inputs, we quantise acoustic representations and find that even 1-bit quantisation can support reasonably high-quality speech.

cs.SD↗