Aholab is the short name of the Signal Processing Laboratory of the University of the Basque Country (UPV/EHU). The laboratory is located in Bilbao. We are a university research team and focus our research in the areas of Text to Speech Conversion, Speech and Speaker Recognition, and Speech Processing in general. Since 2005 we are a recognized research group of the Basque Research Network. The laboratory is part of the Basque Center for Language Technology (HiTZ) and the Department of Communications Engineering of the Faculty of Engineering of Bilbao (ETSI).
Contract duration: 16 months
Available positions: 2
Job Description: We are looking for two people from the IT or telecommunications field, with a passion for artificial intelligence and neural networks. In this exciting project, you will work on developing and refining voice creation and recognition systems.
Responsibilities:
Requirements:
Benefits:
If you are interested in joining our team and contributing to the future of speech recognition technology, we look forward to your application! Tell us!
More information here and here.
From the Aholab Group and the HiTZ center we present to the public the new speech recognition system in Basque. This technological advance has the potential to transform the interaction between people and technology, especially in the field of the Basque language.
The system has been trained with 548 hours of Basque voices from different public sources (Mozilla Common Voice 16.1, Basque Parliament, OpenSLR), which allows it to accurately recognize the words and phrases spoken by users, reaching quality levels of WER less than 5%.
Two different models have been created based on NVIDIA pre-trained models. One of them using a language model with more classic techniques, and the other using more emerging technologies such as transducers. The training of the models was carried out on the Hyperion system from the DIPC servers.
The system can potentially be integrated into virtual assistants to perform tasks such as sending messages, searching for information or setting reminders. It could also enable the automation of responses to telephone calls, improving efficiency and customer service. And it will facilitate the transcription of audio recordings in Basque.
A demo of the speech recognition system in Basque is available at this link and the models are available at Gaitu-Data. The team invites the community to use it and provide feedback to continue improving the technology. We hope that it will be a valuable tool for the Basque community and contribute to the strengthening of our language.
Gurekin ikastera etortzea pentsatu dutenei, egiten duguna azaltzen diegu.
Ate irekien jardunaldia – Bilboko Ingeniaritza Eskola – UPV/EHU
In the framework of the DeepRestore project, acquired EMG signals will be converted into speech. Combination with lip-reading modality will also be tested. The main tasks for the candidate will be:
a) taking part in the design and performing of EMG and video recordings
b) to process and prepare the acquired signals
c) to investigate and evaluate different deep learning strategies to decode the signals into speech
d) to document the process and contribute to scientific publications.
The candidate should preferably have a BSc degree in telecommunications engineering, artificial intelligence, computer science or equivalent preferably with a MSc. Degree. Outstanding curriculum vitae, good programming abilities, education in machine learning and experience in programming is necessary. Strong motivation, team working skills, and fluent spoken and written English will be highly appreciated.
The candidate should send an e-mail in English to inma.hernaez@ehu.eus with a CV and a brief description of the applicant particular merits to get the position. All applications will be considered regardless of gender, age, cultural background, nationality or impairments. Open until filled.