Alert button
Picture for Tanel Alumäe

Tanel Alumäe

Alert button

PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings

Mar 04, 2024
Joonas Kalda, Clément Pagés, Ricard Marxer, Tanel Alumäe, Hervé Bredin

Figure 1 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 2 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 3 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Figure 4 for PixIT: Joint Training of Speaker Diarization and Speech Separation from Real-world Multi-speaker Recordings
Viaarxiv icon

Dialect Adaptation and Data Augmentation for Low-Resource ASR: TalTech Systems for the MADASR 2023 Challenge

Oct 26, 2023
Tanel Alumäe, Jiaming Kong, Daniil Robnikov

Viaarxiv icon

Collar-aware Training for Streaming Speaker Change Detection in Broadcast Speech

May 14, 2022
Joonas Kalda, Tanel Alumäe

Figure 1 for Collar-aware Training for Streaming Speaker Change Detection in Broadcast Speech
Figure 2 for Collar-aware Training for Streaming Speaker Change Detection in Broadcast Speech
Figure 3 for Collar-aware Training for Streaming Speaker Change Detection in Broadcast Speech
Figure 4 for Collar-aware Training for Streaming Speaker Change Detection in Broadcast Speech
Viaarxiv icon

Pretraining Approaches for Spoken Language Recognition: TalTech Submission to the OLR 2021 Challenge

May 14, 2022
Tanel Alumäe, Kunnar Kukk

Figure 1 for Pretraining Approaches for Spoken Language Recognition: TalTech Submission to the OLR 2021 Challenge
Figure 2 for Pretraining Approaches for Spoken Language Recognition: TalTech Submission to the OLR 2021 Challenge
Figure 3 for Pretraining Approaches for Spoken Language Recognition: TalTech Submission to the OLR 2021 Challenge
Figure 4 for Pretraining Approaches for Spoken Language Recognition: TalTech Submission to the OLR 2021 Challenge
Viaarxiv icon

Improving Language Identification of Accented Speech

Apr 01, 2022
Kunnar Kukk, Tanel Alumäe

Figure 1 for Improving Language Identification of Accented Speech
Figure 2 for Improving Language Identification of Accented Speech
Figure 3 for Improving Language Identification of Accented Speech
Figure 4 for Improving Language Identification of Accented Speech
Viaarxiv icon

Robust Training of Vector Quantized Bottleneck Models

May 18, 2020
Adrian Łańcucki, Jan Chorowski, Guillaume Sanchez, Ricard Marxer, Nanxin Chen, Hans J. G. A. Dolfing, Sameer Khurana, Tanel Alumäe, Antoine Laurent

Figure 1 for Robust Training of Vector Quantized Bottleneck Models
Figure 2 for Robust Training of Vector Quantized Bottleneck Models
Figure 3 for Robust Training of Vector Quantized Bottleneck Models
Figure 4 for Robust Training of Vector Quantized Bottleneck Models
Viaarxiv icon

Advanced Rich Transcription System for Estonian Speech

Jan 11, 2019
Tanel Alumäe, Ottokar Tilk, Asadullah

Figure 1 for Advanced Rich Transcription System for Estonian Speech
Figure 2 for Advanced Rich Transcription System for Estonian Speech
Figure 3 for Advanced Rich Transcription System for Estonian Speech
Figure 4 for Advanced Rich Transcription System for Estonian Speech
Viaarxiv icon

Weakly Supervised Training of Speaker Identification Models

Jun 22, 2018
Martin Karu, Tanel Alumäe

Figure 1 for Weakly Supervised Training of Speaker Identification Models
Figure 2 for Weakly Supervised Training of Speaker Identification Models
Figure 3 for Weakly Supervised Training of Speaker Identification Models
Figure 4 for Weakly Supervised Training of Speaker Identification Models
Viaarxiv icon

Low-Resource Neural Headline Generation

Jul 31, 2017
Ottokar Tilk, Tanel Alumäe

Figure 1 for Low-Resource Neural Headline Generation
Figure 2 for Low-Resource Neural Headline Generation
Figure 3 for Low-Resource Neural Headline Generation
Figure 4 for Low-Resource Neural Headline Generation
Viaarxiv icon