News

Cool Timeline

2021
The LIUM Human Active Correction Platform for Speaker Diarization

The human assisted speaker diarization platform enables a human annotator to correct the output of any speaker diarization system by providing a graphical view of the diarization segmentation and clustering steps and guiding the human annotator to optimize the correction process and easily improve the resulting diarization. This plateform was…Lire la suite

New publications in ICASSP and IberSpeech

ICASSP 2021: Speaker Embedding for Diarization of Broadcast data in the ALLIES Challenge, Anthony Larcher, Ambuj Mehrish, Marie Tahon, Sylvain Meignier, Jean Carrive, David Doukhan, Olivier Galibert, Nicholas Evans IberSpeech 2021: Active correction for speaker diarization with human in the loop, Yevhenii Prokopalo, Meysam Shamsi, Loïc Barrault, Sylvain Meignier, Anthony…Lire la suite

2020
Presentation of ALLIES Evaluation at EMNLP

WMT_LLMT: Loïc Barrault presented Lifelong Learning for Machine Translation Evaluation at the workshop WMT (http://statmt.org/wmt20/), which took place during EMNLP conference (https://2020.emnlp.org/) Barrault L., Bougares F., Costa-Jussa M., Biesialska M., Galibert O., A Shared Task on Lifelong Learning Machine Translation, 2020

Presentation of ALLIES Evaluation at EAMT

Loïc Barrault presented Lifelong Learning for Machine Translation Evaluation during WMT_LLMT_EAMT (Lisbon, Portugal): https://eamt2020.inesc-id.pt/ Barrault L., Bougares F., Costa-Jussa M., Biesialska M., Galibert O., A Shared Task on Lifelong Learning Machine Translation, 2020

ALLIES evaluation

Allies first batch of evaluation data is ready for speaker Diarization

Arrival of Meysam SHAMSI

Allies research team is welcoming Meysam SHAMSI who will work on human assisted learning for speaker diarization.

evallies python module

Release of the evallies python module for lifelong speaker diarization

ALLIES evaluation

Allies machine translation évaluation is now running

LREC publication for Yevhenii Prokopalo

Yevhenii Prokopalo, Sylvain Meignier, Olivier Galibert, Loïc Barrault, Anthony Larcher. Evaluation of Lifelong Learning Systems. International Conference on Language Resources and Evaluation, May 2020, Marseille, France. hal-02496971

Welcome in the speaker diarization challenge

Welcome to the first teams of participants in the speaker diarization challenge

ALLIES evaluation: Registration open

Registration for ALLIES evaluation is now open.

2019
ALLIES Evaluation for Autonomous Speaker Diarization Systems

ALLIES technical documentation ALLIES evaluation plan description of the task ALLIES baseline system and tutorial the complete ALLIES documentation Documentation of the BEAT platform more information on the BEAT platform Conda environment file description required to setup your ALLIES environment Beat prefix copy of the baseline diarization system (not required…Lire la suite

Delivery of diarization data by INA

200 hours of audio-video data to be used for the Albayzin-ALLIES evaluation. Annotation is in progress.  

New version of BEAT-CORE (IDIAP)

Integration of user simulation loops for the evaluation of active and interactive learning processes

ALLIES meeting in Barcelona

ALLIES meeting at Universitat politècnica de Catalunya, Barcelona, Spain.

Hackathon in Martigny (IDIAP)

Members of ALLIES consortium participated in the hackathon, held in Martigny, Switzerland from 8 to 12 July 2019

CHIST-ERA conference in Tallinn

ALLIES consortium participated in the CHIST-ERA conference, held in Tallinn, Estonia on 10 and 11 June 2019

ALLIES meeting in Bucharest

Meeting in Bucharest, Romania Presentation of ALLIES project during the annual CHIST-ERA meeting

2018
ALLIES meeting in Martigny

ALLIES meeting at Idiap Research Institute, Martigny, Switzerland