Named Entity Recognition for Spoken Finnish

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

4 Citations (Scopus)
123 Downloads (Pure)


In this paper we present a Bidirectional LSTM neural network with a Conditional Random Field layer on top, which utilizes word, character and morph embeddings in order to perform named entity recognition on various Finnish datasets. To overcome the lack of annotated training corpora that arises when dealing with low-resource languages like Finnish, we tried a knowledge transfer technique to transfer tags from Estonian dataset. On the human annotated in-domain Digitoday dataset, out system achieved F1 score of 84.73. On the out-of-domain Wikipedia set we got F1 score of 67.66. In order to see how well the system performs on speech data, we used two datasets containing automatic speech recognition outputs. Since we do not have true labels for those datasets, we used a rule-based system to annotate them and used those annotations as reference labels. On the first dataset which contains Finnish parliament sessions we obtained F1 score of 42.09 and on the second one which contains talks from Yle Pressiklubi we obtained F1 score of 74.54.

Original languageEnglish
Title of host publicationAI4TV 2020 - Proceedings of the 2nd International Workshop on AI for Smart TV Content Production, Access and Delivery
Number of pages5
ISBN (Electronic)9781450381468
Publication statusPublished - 12 Oct 2020
MoE publication typeA4 Conference publication
EventInternational Workshop on AI for Smart TV Content Production, Access and Delivery - Virtual, Online, United States
Duration: 12 Oct 202012 Oct 2020
Conference number: 2


WorkshopInternational Workshop on AI for Smart TV Content Production, Access and Delivery
Abbreviated titleAI4TV
Country/TerritoryUnited States
CityVirtual, Online


  • low-resource
  • named entity recognition
  • speech recognition


Dive into the research topics of 'Named Entity Recognition for Spoken Finnish'. Together they form a unique fingerprint.

Cite this