We present our IWSLT 2025 submission for the low-resource track on North Levantine Arabic to English speech translation, building on our IWSLT 2024 efforts. We retain last year’s cascade ASR architecture that combines a TDNN-F model and a Zipformer for the ASR step. We upgrade the Zipformer to the Zipformer-Large variant (253 M parameters vs. 66 M) to capture richer acoustic representations. For the MT part, to further alleviate data sparsity, we created a crowd-sourced parallel corpus covering five major Arabic dialects (Tunisian, Levantine, Moroccan, Algerian, Egyptian) curated via rigorous qualification and filtering. We show that using crowd-sourced data is feasible in low-resource scenarios as we observe improved automatic evaluation metrics across all dialects. We also experimented with the dataset under a high-resource scenario, where we had access to a large, high-quality Levantine Arabic corpus from LDC. In this setting, adding the crowd-sourced data does not improve the scores on the official validation set anymore. Our final submission scores 20.0 BLEU on the official test set.

This paper presents ALADAN’s approach to the IWSLT 2024 Dialectal and Low-resource shared task, focusing on Levantine Arabic (apc) and Tunisian Arabic (aeb) to English speech translation (ST). Addressing challenges such as the lack of standardized orthography and limited training data, we propose a solution for data normalization in Dialectal Arabic, employing a modified Levenshtein distance and Word2vec models to find orthographic variants of the same word. Our system consists of a cascade ST system integrating two ASR systems (TDNN-F and Zipformer) and two NMT modules derived from pre-trained models (NLLB-200 1.3B distilled model and CohereAI’s Command-R). Additionally, we explore the integration of unsupervised textual and audio data, highlighting the importance of multi-dialectal datasets for both ASR and NMT tasks. Our system achieves BLEU score of 31.5 for Levantine Arabic on the official validation set.

This paper documents the systems developed by LIMSI for the IWSLT 2014 speech translation task (English→French). The main objective of this participation was twofold: adapting different components of the ASR baseline system to the peculiarities of TED talks and improving the machine translation quality on the automatic speech recognition output data. For the latter task, various techniques have been considered: punctuation and number normalization, adaptation to ASR errors, as well as the use of structured output layer neural network models for speech data.

This paper describes the speech-to-text systems used to provide automatic transcriptions used in the Quaero 2010 evaluation of Machine Translation from speech. Quaero (www.quaero.org) is a large research and industrial innovation program focusing on technologies for automatic analysis and classification of multimedia and multilingual documents. The ASR transcript is the result of a Rover combination of systems from three teams ( KIT, RWTH, LIMSI+VR) for the French and German languages. The casesensitive word error rates (WER) of the combined systems were respectively 20.8% and 18.1% on the 2010 evaluation data, relative WER reductions of 14.6% and 17.4% respectively over the best component system.

Being the client’s first interface, call centres worldwide contain a huge amount of information of all kind under the form of conversational speech. If accessible, this information can be used to detect eg. major events and organizational flaws, improve customer relations and marketing strategies. An efficient way to exploit the unstructured data of telephone calls is data-mining, but current techniques apply on text only. The CallSurf project gathers a number of academic and industrial partners covering the complete platform, from automatic transcription to information retrieval and data mining. This paper concentrates on the speech recognition module as it discusses the collection, the manual transcription of the training corpus and the techniques used to build the language model. The NLP techniques used to pre-process the transcribed corpus for data mining are POS tagging, lemmatization, noun group and named entity recognition. Some of them have been especially adapted to the conversational speech characteristics. POS tagging and preliminary data mining results obtained on the manually transcribed corpus are briefly discussed.