English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Multi-source localization is an important and challenging technique formulti-talker conversation analysis. This paper proposes a novel supervisedlearning method using deep neural networks to estimate the direction of arrival(DOA) of all the speakers simultaneously from the audio mixture. At the heartof the proposal is a source splitting mechanism that creates source-specificintermediate representations inside the network. This allows our model to givesource-specific posteriors as the output unlike the traditional multi-labelclassification approach. Existing deep learning methods perform a frame levelprediction, whereas our approach performs an utterance level prediction byincorporating temporal selection and averaging inside the network to avoidpost-processing. We also experiment with various loss functions and show that avariant of earth mover distance (EMD) is very effective in classifying DOA at avery high resolution by modeling inter-class relationships. In addition tousing the prediction error as a metric for evaluating our localization model,we also establish its potency as a frontend with automatic speech recognition(ASR) as the downstream task. We convert the estimated DOAs into a featuresuitable for ASR and pass it as an additional input feature to a strongmulti-channel and multi-talker speech recognition baseline. This added inputfeature drastically improves the ASR performance and gives a word error rate(WER) of 6.3% on the evaluation data of our simulated noisy two speakermixtures, while the baseline which doesn't use explicit localization input hasa WER of 11.5%. We also perform ASR evaluation on real recordings with theoverlapped set of the MC-WSJ-AV corpus in addition to simulated mixtures.

Deep Learning based Multi-Source Localization with Source Splitting and its Effectiveness in Multi-Talker Speech Recognition