{"url":"/task/speaker-diarization","name":"Speaker Diarization","slug":"speaker-diarization","description_markdown":"**Speaker Diarization** is the task of segmenting and co-indexing audio recordings by speaker. The way the task is commonly defined, the goal is not to identify known speakers, but to co-index segments that are attributed to the same speaker; in other words, diarization implies finding speaker boundaries and grouping segments that belong to the same speaker, and, as a by-product, determining the number of distinct speakers. In combination with speech recognition, diarization enables speaker-attributed speech-to-text transcription.\n\n\n<span class=\"description-source\">Source: [Improving Diarization Robustness using Diversification, Randomization and the DOVER Algorithm ](https://arxiv.org/abs/1910.11691)</span>","categories":[{"name":"Speech","url":"/area/speech"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":328,"papers_with_code":93,"benchmarks":12,"benchmark_tables_in_archive":15,"benchmark_tables_shown":15,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":11,"subtasks":0,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/speaker-diarization-on-callhome","slug":"speaker-diarization-on-callhome","dataset":"CALLHOME","dataset_url":"/dataset/callhome-american-english-speech","rows_in_archive":10,"metrics":["DER(%)","DER(ig olp)","FA","MI","CF"],"first_row_in_archive_order":{"model":"TOLD","paper_title":"TOLD: A Novel Two-Stage Overlap-Aware Framework for Speaker Diarization","paper_url":"/paper/told-a-novel-two-stage-overlap-aware","paper_date":"2023-03-08","arxiv_id":"2303.05397","code_links":[{"title":"alibaba-damo-academy/FunASR","url":"https://github.com/alibaba-damo-academy/FunASR"}],"syntology":null}},{"leaderboard":"/sota/speaker-diarization-on-nist-sre-2000","slug":"speaker-diarization-on-nist-sre-2000","dataset":"NIST-SRE 2000","dataset_url":null,"rows_in_archive":5,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"x-vector (MCGAN)","paper_title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","paper_url":"/paper/titanet-neural-model-for-speaker","paper_date":"2021-10-08","arxiv_id":"2110.04410","code_links":[{"title":"NVIDIA/NeMo","url":"https://github.com/NVIDIA/NeMo"},{"title":"Wadaboa/titanet","url":"https://github.com/Wadaboa/titanet"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-ami-lapel","slug":"speaker-diarization-on-ami-lapel","dataset":"AMI Lapel","dataset_url":null,"rows_in_archive":4,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"TitaNet-M (NME-SC)","paper_title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","paper_url":"/paper/titanet-neural-model-for-speaker","paper_date":"2021-10-08","arxiv_id":"2110.04410","code_links":[{"title":"NVIDIA/NeMo","url":"https://github.com/NVIDIA/NeMo"},{"title":"Wadaboa/titanet","url":"https://github.com/Wadaboa/titanet"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-ami-mixheadset","slug":"speaker-diarization-on-ami-mixheadset","dataset":"AMI MixHeadset","dataset_url":null,"rows_in_archive":4,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"TitaNet-L (NME-SC)","paper_title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","paper_url":"/paper/titanet-neural-model-for-speaker","paper_date":"2021-10-08","arxiv_id":"2110.04410","code_links":[{"title":"NVIDIA/NeMo","url":"https://github.com/NVIDIA/NeMo"},{"title":"Wadaboa/titanet","url":"https://github.com/Wadaboa/titanet"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-ch109","slug":"speaker-diarization-on-ch109","dataset":"CH109","dataset_url":null,"rows_in_archive":4,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"TitaNet-S (NME-SC)","paper_title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","paper_url":"/paper/titanet-neural-model-for-speaker","paper_date":"2021-10-08","arxiv_id":"2110.04410","code_links":[{"title":"NVIDIA/NeMo","url":"https://github.com/NVIDIA/NeMo"},{"title":"Wadaboa/titanet","url":"https://github.com/Wadaboa/titanet"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-dihard-1","slug":"speaker-diarization-on-dihard-1","dataset":"DIHARD","dataset_url":null,"rows_in_archive":3,"metrics":["DER(%)","FA","Miss"],"first_row_in_archive_order":{"model":"pyannote (waveform)","paper_title":"pyannote.audio: neural building blocks for speaker diarization","paper_url":"/paper/pyannoteaudio-neural-building-blocks-for","paper_date":"2019-11-04","arxiv_id":"1911.01255","code_links":[{"title":"pyannote/pyannote-audio","url":"https://github.com/pyannote/pyannote-audio"},{"title":"muskang48/Speaker-Diarization","url":"https://github.com/muskang48/Speaker-Diarization"},{"title":"MarvinLvn/voice-type-classifier","url":"https://github.com/MarvinLvn/voice-type-classifier"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-etape","slug":"speaker-diarization-on-etape","dataset":"ETAPE","dataset_url":null,"rows_in_archive":3,"metrics":["DER(%)","FA","Miss"],"first_row_in_archive_order":{"model":"pyannote (waveform)","paper_title":"pyannote.audio: neural building blocks for speaker diarization","paper_url":"/paper/pyannoteaudio-neural-building-blocks-for","paper_date":"2019-11-04","arxiv_id":"1911.01255","code_links":[{"title":"pyannote/pyannote-audio","url":"https://github.com/pyannote/pyannote-audio"},{"title":"muskang48/Speaker-Diarization","url":"https://github.com/muskang48/Speaker-Diarization"},{"title":"MarvinLvn/voice-type-classifier","url":"https://github.com/MarvinLvn/voice-type-classifier"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-ami","slug":"speaker-diarization-on-ami","dataset":"AMI","dataset_url":null,"rows_in_archive":2,"metrics":["DER(%)","FA","Miss"],"first_row_in_archive_order":{"model":"pyannote (waveform)","paper_title":"pyannote.audio: neural building blocks for speaker diarization","paper_url":"/paper/pyannoteaudio-neural-building-blocks-for","paper_date":"2019-11-04","arxiv_id":"1911.01255","code_links":[{"title":"pyannote/pyannote-audio","url":"https://github.com/pyannote/pyannote-audio"},{"title":"muskang48/Speaker-Diarization","url":"https://github.com/muskang48/Speaker-Diarization"},{"title":"MarvinLvn/voice-type-classifier","url":"https://github.com/MarvinLvn/voice-type-classifier"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-callhome-109","slug":"speaker-diarization-on-callhome-109","dataset":"CALLHOME-109","dataset_url":"/dataset/callhome-american-english-speech","rows_in_archive":2,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"titanet-s","paper_title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","paper_url":"/paper/titanet-neural-model-for-speaker","paper_date":"2021-10-08","arxiv_id":"2110.04410","code_links":[{"title":"NVIDIA/NeMo","url":"https://github.com/NVIDIA/NeMo"},{"title":"Wadaboa/titanet","url":"https://github.com/Wadaboa/titanet"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/speaker-diarization-on-alimeeting","slug":"speaker-diarization-on-alimeeting","dataset":"AliMeeting","dataset_url":"/dataset/alimeeting","rows_in_archive":1,"metrics":["DER(%)"],"first_row_in_archive_order":{"model":"SOND","paper_title":"Speaker Embedding-aware Neural Diarization: an Efficient Framework for Overlapping Speech Diarization in Meeting Scenarios","paper_url":"/paper/speaker-embedding-aware-neural-diarization-a","paper_date":"2022-03-18","arxiv_id":"2203.09767","code_links":[{"title":"alibaba-damo-academy/FunASR","url":"https://github.com/alibaba-damo-academy/FunASR"}],"syntology":null}},{"leaderboard":"/sota/speaker-diarization-on-dihard-ii","slug":"speaker-diarization-on-dihard-ii","dataset":"DIHARD II","dataset_url":"/dataset/dihard-ii","rows_in_archive":1,"metrics":["DER(%)","DER - no overlap"],"first_row_in_archive_order":{"model":"UIS-RNN-SML","paper_title":"Supervised online diarization with sample mean loss for multi-domain data","paper_url":"/paper/supervised-online-diarization-with-sample","paper_date":"2019-11-04","arxiv_id":"1911.01266","code_links":[{"title":"DonkeyShot21/uis-rnn-sml","url":"https://github.com/DonkeyShot21/uis-rnn-sml"}],"syntology":null}},{"leaderboard":"/sota/speaker-diarization-on-hub5-00-callhome","slug":"speaker-diarization-on-hub5-00-callhome","dataset":"Hub5'00 CallHome","dataset_url":"/dataset/callhome-american-english-speech","rows_in_archive":1,"metrics":["V"],"first_row_in_archive_order":{"model":"UIS-RNN","paper_title":"Fully Supervised Speaker Diarization","paper_url":"/paper/fully-supervised-speaker-diarization","paper_date":"2018-10-10","arxiv_id":"1810.04719","code_links":[{"title":"google/uis-rnn","url":"https://github.com/google/uis-rnn"}],"syntology":null}},{"leaderboard":null,"slug":"speaker-diarization-on-call-home-american","dataset":"call_home_american_english_speech","dataset_url":null,"rows_in_archive":0,"metrics":["Test DER"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"speaker-diarization-on-callhome-nist-sre-2000","dataset":"CALLHOME (NIST-SRE-2000 Disc8)","dataset_url":null,"rows_in_archive":0,"metrics":["Test DER"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"speaker-diarization-on-dihard3-eval","dataset":"DIHARD3-eval","dataset_url":null,"rows_in_archive":0,"metrics":["Test DER"],"first_row_in_archive_order":null}],"datasets":[{"url":"/dataset/ava","name":"AVA","full_name":"Atomic Visual Actions","num_papers_in_archive":113},{"url":"/dataset/alimeeting","name":"AliMeeting","full_name":"Multi-Channel Multi-Party Meeting Transcription Challenge","num_papers_in_archive":46},{"url":"/dataset/chime-5","name":"CHiME-5","full_name":"CHiME Speech Separation and Recognition Challenge","num_papers_in_archive":42},{"url":"/dataset/dihard-ii","name":"DIHARD II","full_name":"","num_papers_in_archive":34},{"url":"/dataset/ava-activespeaker","name":"AVA-ActiveSpeaker","full_name":"","num_papers_in_archive":22},{"url":"/dataset/magicdata-ramc","name":"MagicData-RAMC","full_name":"","num_papers_in_archive":12},{"url":"/dataset/callhome-american-english-speech","name":"CALLHOME American English Speech","full_name":"","num_papers_in_archive":11},{"url":"/dataset/ava-speech","name":"AVA-Speech","full_name":"","num_papers_in_archive":10},{"url":"/dataset/asr-ramc-bigccsc-a-chinese-conversational","name":"ASR-RAMC-BIGCCSC: A CHINESE CONVERSATIONAL SPEECH CORPUS","full_name":"","num_papers_in_archive":1},{"url":"/dataset/fsc-p2","name":"FSC-P2","full_name":"Fearless Steps Challenge Phase2","num_papers_in_archive":1},{"url":"/dataset/radiotalk","name":"RadioTalk","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":93,"tagged_in_all":328,"items":[{"url":"/paper/ava-avd-audio-visual-speaker-diarization-in","title":"AVA-AVD: Audio-Visual Speaker Diarization in the Wild","date":"2021-11-29","arxiv_id":"2111.14448","repositories_listed":9,"syntology":null},{"url":"/paper/ava-avd-audio-visual-speaker-diarization-in","title":"AVA-AVD: Audio-Visual Speaker Diarization in the Wild","date":"2021-11-29","arxiv_id":"2111.14448","repositories_listed":9,"syntology":null},{"url":"/paper/speaker-diarization-with-lstm","title":"Speaker Diarization with LSTM","date":"2017-10-28","arxiv_id":"1710.10468","repositories_listed":4,"syntology":null},{"url":"/paper/speaker-diarization-with-lstm","title":"Speaker Diarization with LSTM","date":"2017-10-28","arxiv_id":"1710.10468","repositories_listed":4,"syntology":null},{"url":"/paper/speech-emotion-diarization-which-emotion","title":"Speech Emotion Diarization: Which Emotion Appears When?","date":"2023-06-22","arxiv_id":"2306.12991","repositories_listed":3,"syntology":null},{"url":"/paper/speech-emotion-diarization-which-emotion","title":"Speech Emotion Diarization: Which Emotion Appears When?","date":"2023-06-22","arxiv_id":"2306.12991","repositories_listed":3,"syntology":null},{"url":"/paper/the-third-dihard-diarization-challenge","title":"The Third DIHARD Diarization Challenge","date":"2020-12-02","arxiv_id":"2012.01477","repositories_listed":3,"syntology":null},{"url":"/paper/the-third-dihard-diarization-challenge","title":"The Third DIHARD Diarization Challenge","date":"2020-12-02","arxiv_id":"2012.01477","repositories_listed":3,"syntology":null},{"url":"/paper/end-to-end-speaker-diarization-for-an-unknown","title":"End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractors","date":"2020-05-20","arxiv_id":"2005.09921","repositories_listed":3,"syntology":{"n":21,"n_ran":0,"n_unverified":21,"n_pointer_only":0}},{"url":"/paper/end-to-end-speaker-diarization-for-an-unknown","title":"End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractors","date":"2020-05-20","arxiv_id":"2005.09921","repositories_listed":3,"syntology":{"n":21,"n_ran":0,"n_unverified":21,"n_pointer_only":0}},{"url":"/paper/speech-recognition-and-multi-speaker","title":"Speech Recognition and Multi-Speaker Diarization of Long Conversations","date":"2020-05-16","arxiv_id":"2005.08072","repositories_listed":3,"syntology":null},{"url":"/paper/speech-recognition-and-multi-speaker","title":"Speech Recognition and Multi-Speaker Diarization of Long Conversations","date":"2020-05-16","arxiv_id":"2005.08072","repositories_listed":3,"syntology":null},{"url":"/paper/pyannoteaudio-neural-building-blocks-for","title":"pyannote.audio: neural building blocks for speaker diarization","date":"2019-11-04","arxiv_id":"1911.01255","repositories_listed":3,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/pyannoteaudio-neural-building-blocks-for","title":"pyannote.audio: neural building blocks for speaker diarization","date":"2019-11-04","arxiv_id":"1911.01255","repositories_listed":3,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/3d-speaker-toolkit-an-open-source-toolkit-for","title":"3D-Speaker-Toolkit: An Open-Source Toolkit for Multimodal Speaker Verification and Diarization","date":"2024-03-29","arxiv_id":"2403.19971","repositories_listed":2,"syntology":null},{"url":"/paper/3d-speaker-toolkit-an-open-source-toolkit-for","title":"3D-Speaker-Toolkit: An Open-Source Toolkit for Multimodal Speaker Verification and Diarization","date":"2024-03-29","arxiv_id":"2403.19971","repositories_listed":2,"syntology":null},{"url":"/paper/diarizationlm-speaker-diarization-post","title":"DiarizationLM: Speaker Diarization Post-Processing with Large Language Models","date":"2024-01-07","arxiv_id":"2401.03506","repositories_listed":2,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/diarizationlm-speaker-diarization-post","title":"DiarizationLM: Speaker Diarization Post-Processing with Large Language Models","date":"2024-01-07","arxiv_id":"2401.03506","repositories_listed":2,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/self-supervised-audio-teacher-student","title":"Self-supervised Audio Teacher-Student Transformer for Both Clip-level and Frame-level Tasks","date":"2023-06-07","arxiv_id":"2306.04186","repositories_listed":2,"syntology":null},{"url":"/paper/self-supervised-audio-teacher-student","title":"Self-supervised Audio Teacher-Student Transformer for Both Clip-level and Frame-level Tasks","date":"2023-06-07","arxiv_id":"2306.04186","repositories_listed":2,"syntology":null},{"url":"/paper/ber-balanced-error-rate-for-speaker","title":"BER: Balanced Error Rate For Speaker Diarization","date":"2022-11-08","arxiv_id":"2211.04304","repositories_listed":2,"syntology":null},{"url":"/paper/ber-balanced-error-rate-for-speaker","title":"BER: Balanced Error Rate For Speaker Diarization","date":"2022-11-08","arxiv_id":"2211.04304","repositories_listed":2,"syntology":null},{"url":"/paper/paddlespeech-an-easy-to-use-all-in-one-speech-1","title":"PaddleSpeech: An Easy-to-Use All-in-One Speech Toolkit","date":"2022-05-20","arxiv_id":"2205.12007","repositories_listed":2,"syntology":null},{"url":"/paper/from-simulated-mixtures-to-simulated","title":"From Simulated Mixtures to Simulated Conversations as Training Data for End-to-End Neural Diarization","date":"2022-04-02","arxiv_id":"2204.00890","repositories_listed":2,"syntology":null},{"url":"/paper/from-simulated-mixtures-to-simulated","title":"From Simulated Mixtures to Simulated Conversations as Training Data for End-to-End Neural Diarization","date":"2022-04-02","arxiv_id":"2204.00890","repositories_listed":2,"syntology":null},{"url":"/paper/speaker-embedding-aware-neural-diarization","title":"Speaker Embedding-aware Neural Diarization for Flexible Number of Speakers with Textual Information","date":"2021-11-28","arxiv_id":"2111.13694","repositories_listed":2,"syntology":null},{"url":"/paper/bertraffic-a-robust-bert-based-approach-for","title":"BERTraffic: BERT-based Joint Speaker Role and Speaker Change Detection for Air Traffic Control Communications","date":"2021-10-12","arxiv_id":"2110.05781","repositories_listed":2,"syntology":null},{"url":"/paper/bertraffic-a-robust-bert-based-approach-for","title":"BERTraffic: BERT-based Joint Speaker Role and Speaker Change Detection for Air Traffic Control Communications","date":"2021-10-12","arxiv_id":"2110.05781","repositories_listed":2,"syntology":null},{"url":"/paper/titanet-neural-model-for-speaker","title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","date":"2021-10-08","arxiv_id":"2110.04410","repositories_listed":2,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/titanet-neural-model-for-speaker","title":"TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context","date":"2021-10-08","arxiv_id":"2110.04410","repositories_listed":2,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}],"syntology_records":8,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}