Empowering Large Language Models for End-to-End Speech Translation Leveraging Synthetic Data
The DOI matches its Crossref record.
Abstract
Speech-to-speech translation (S2ST) is a key technology for seamless cross-lingual communication. Traditional cascaded systems, which involve speech recognition, text translation, and speech synthesis, are prone to error propagation and latency. In this work, we present SLAM-TR, an end-to-end speech translation model which directly maps input speech to output speech, eliminating the need for intermediate text representations. By fine-tuning from the large language model Qwen2-0.5B, SLAM-TR achieves superior performance over the cascaded baseline and state-of-the-art open-source models with minimal training time. Additionally, SLAM-TR demonstrates strong generalization, achieving an ASR-BLEU score of 8.2 on the FLEURS benchmark, outperforming both cascaded and open-source systems. Addressing the challenge of limited natural speech translation data, we propose SynStard-1000, a 1,000-hour synthetic speech translation dataset. We create the dataset by synthesizing speech data from high-quality parallel text pairs. And the open-source release of this dataset contributes to the field of speech-to-speech translation.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Prior studies have demonstrated the technical feasibility of text-based VSPs and AI-assisted automated scoring systems.”
Citations
Cites
Still being checked.
Cited by
Chen, S.-C., Tsai, I.-N., Lin, Y.-C., Huang, S.-F., & Lee, H.-Y. (2026) · Interspeech (USB) · doi:10.21437/interspeech.2026-42 · Cited by 3
Thesis Proposal: Development of End-to-End Speech Translation Models for Indian Languages
Jamaluddin (2026) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2026.eacl-srw.41 · Cited by 2
Central Kurdish Text-to-Speech and Its Application in Speech-to-Text Translation
Mohammadamini, M., Shamsi, M., & Tahon, M. (2026) · Language Resources and Evaluation Conference (LREC) · doi:10.63317/4hfwowidu34u · Cited by 2
Maltais, M., Jeon, Y., Ma, M., Muhammad, S. H., Abdulmumin, I., Mukhtar, M. I., Abolade, D. O., Okepefi, J., Sewedo, J., & Adelani, D. I. (2026) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/2026.emnlp-main.665 · Cited by 1
Song, J. W., Kim, M., Hong, C., Kim, Y. S., Cho, J., Kim, J. H., Myung, J., Choi, A., Yoon, H., Lee, S. G. W., et al. (2026) · medRxiv · doi:10.64898/2026.02.21.26346803
Metadata
- Authors
- Pu, Y., Liu, X., Zhang, G., Yan, Z., Zhang, W.-Q., & Chen, X.
- Venue
- Interspeech 2025
- Publisher
- ISCA
- Published
- 2025-08-17
- Pages
- 26-30
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W4415432961
- Cited by
- 1
- References
- 0
- Citations by year
- 2026: 1
- FWCI
- 0.95
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Interspeech (USB) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- end-to-end speech translation, large language models, synthetic data, data augmentation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.