Joan Albert Silvestre-Cerdà

Older adults often need guidance when visiting new buildings for the first time. However, indoor navigation remains challenging due to the lack of Global Positioning System (GPS) availability, visually repetitive corridors, and frequent location failures. This article presents a multimodal indoor navigation assistant that combines graph-based route planning with visual landmark verification to provide step-by-step guidance. The environment is modelled as a directed graph whose nodes are annotated with semantic landmarks, and the graph is constructed primarily from a video of the building, reducing the need for 3D scanners, beacons, or other specialised instruments. Routes are calculated using Dijkstra’s shortest-path algorithm over the semantic graph. During navigation, camera frames are analysed using a restricted vision-language recognition strategy that only considers candidate landmarks from the current and next nodes, reducing false detections and improving interpretability. To increase robustness, a temporary voting mechanism was introduced to confirm node transitions, as well as a hierarchical redirection strategy with local and global recovery. The system is implemented in two modes: handheld mode with visual cues using augmented reality arrows, mini map and voice instructions, and hands-free mode with front camera using voice instructions and keywords. Evaluation involved preliminary technical testing in the United Kingdom followed by formal user validation in Spain. During these trials, participants reported high usability, strong confidence and safety, and increased perceived independence.

This paper describes the automatic speech recognition (ASR) systems built by the MLLP-VRAIN research group of Universitat Politècnica de València for the Albayzín-RTVE 2020 Speech-to-Text Challenge, and includes an extension of the work consisting of building and evaluating equivalent systems under the closed data conditions from the 2018 challenge. The primary system (p-streaming_1500ms_nlt) was a hybrid ASR system using streaming one-pass decoding with a context window of 1.5 seconds. This system achieved 16.0% WER on the test-2020 set. We also submitted three contrastive systems. From these, we highlight the system c2-streaming_600ms_t which, following a similar configuration as the primary system with a smaller context window of 0.6 s, scored 16.9% WER points on the same test set, with a measured empirical latency of 0.81 ± 0.09 s (mean ± stdev). That is, we obtained state-of-the-art latencies for high-quality automatic live captioning with a small WER degradation of 6% relative. As an extension, the equivalent closed-condition systems obtained 23.3% WER and 23.5% WER, respectively. When evaluated with an unconstrained language model, we obtained 19.9% WER and 20.4% WER; i.e., not far behind the top-performing systems with only 5% of the full acoustic data and with the extra ability of being streaming-capable. Indeed, all of these streaming systems could be put into production environments for automatic captioning of live media streams.

Although Long-Short Term Memory (LSTM) networks and deep Transformers are now extensively used in offline ASR, it is unclear how best offline systems can be adapted to work with them under the streaming setup. After gaining considerable experience on this regard in recent years, in this paper we show how an optimized, low-latency streaming decoder can be built in which bidirectional LSTM acoustic models, together with general interpolated language models, can be nicely integrated with minimal perfomance degradation. In brief, our streaming decoder consists of a one-pass, real-time search engine relying on a limited-duration window sliding over time and a number of ad hoc acoustic and language model pruning techniques. Extensive empirical assessment is provided on truly streaming tasks derived from the well-known LibriSpeech and TED talks datasets, as well as from TV shows on a main Spanish broadcasting station.

Online video lecture repositories are rapidly growing and becoming established as fundamental knowledge assets. However, most lectures are neither transcribed nor translated because of the lack of cost-effective solutions that can give accurate enough results. In this paper, we describe a system architecture that supports the cost-effective transcription and translation of large video lecture repositories. This architecture has been adopted in the EU project transLectures and is now being tested on a repository of more than 9000 video lectures at the Universitat Politecnica de Valencia. Following a brief description of this repository and of the transLectures project, we describe the proposed system architecture in detail. We also report empirical results on the quality of the transcriptions and translations currently being maintained and steadily improved.