- \n
- A collection of multilingual speaker diarization datasets that are compatible with the diarizers library. They have been processed using diarizers scripts. \n
The available datasets are the CallHome (Japanese, Chinese, German, Spanish, English), AMI Corpus (English), Vox-Converse (English) and Simsamu (French). We aim to add more datasets in the future to better support speaker diarization on the Hub.
\n- \n
- A collection of multilingual fine-tuned segmentation model baselines compatible with pyannote. \n
Each model has been fine-tuned on a specific Callhome language subset. They achieve better performances on multilingual data compared to pyannote's pre-trained segmentation-3.0 model (see benchmark for more details on model performance).
\nTogether with diarizers-community, we release:
\n- \n
diarizers, a library for fine-tuning pyannote speaker diarization models using the Hugging Face ecosystem.
\n \nA google colab notebook, with a step-by-step guide on how to use diarizers.
\n \n
Benchmark
\n| Callhome test dataset | \nModel | \nDER | \nFalse alarm | \nMissed detection | \nConfusion | \n
|---|---|---|---|---|---|
| Japanese | \nPretrained | \n25.44 | \n2.30 | \n17.45 | \n5.69 | \n
| \n | Fine-tuned | \n18.23 | \n6.31 | \n6.91 | \n5.01 | \n
| Spanish | \nPretrained | \n33.44 | \n2.59 | \n25.19 | \n5.66 | \n
| \n | Fine-tuned | \n25.72 | \n6.87 | \n12.73 | \n6.12 | \n
| English | \nPretrained | \n22.16 | \n6.29 | \n10.97 | \n4.90 | \n
| \n | Fine-tuned | \n18.40 | \n7.10 | \n6.98 | \n4.32 | \n
| German | \nPretrained | \n21.90 | \n3.10 | \n14.25 | \n4.55 | \n
| \n | Fine-tuned | \n16.75 | \n5.00 | \n7.75 | \n4.00 | \n
| Chinese | \nPretrained | \n19.73 | \n4.81 | \n9.82 | \n5.11 | \n
| \n | Fine-tuned | \n15.95 | \n5.04 | \n7.24 | \n3.68 | \n
Results are in %. They have been obtained using the test script from diarizers.
\n","classNames":"hf-sanitized hf-sanitized-NRGP8XtuaSR8pSWcFNcMM"},"users":[{"_id":"65e5c224a0a2f649dab3dfe4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/60r3E2R2R61oGvM24DSqH.jpeg","isPro":false,"fullname":"Kamil Akesbi","user":"kamilakesbi","type":"user"},{"_id":"61f91cf54a8e5a275b2b3e7c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1653243468328-61f91cf54a8e5a275b2b3e7c.jpeg","isPro":false,"fullname":"Sanchit Gandhi","user":"sanchit-gandhi","type":"user"},{"_id":"61b85ce86eb1f2c5e6233736","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1655385361868-61b85ce86eb1f2c5e6233736.jpeg","isPro":false,"fullname":"Vaibhav Srivastav","user":"reach-vb","type":"user"},{"_id":"6528662d94bb032403cda789","avatarUrl":"/avatars/0f211a81a0f420a773c85fd38a6eaff4.svg","isPro":false,"fullname":"Brian MacWhinney","user":"macwhinney","type":"user"},{"_id":"61cb65a73918092781f7b775","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61cb65a73918092781f7b775/StxBZ3rF4jfQ4c67ZYnzp.png","isPro":false,"fullname":"Houjun Liu","user":"jemoka","type":"user"},{"_id":"5fcf602140b27fad857dfa8d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5fcf602140b27fad857dfa8d/zFcElfPxAsEXYCkfOHflp.png","isPro":false,"fullname":"Hervé Bredin","user":"hbredin","type":"user"},{"_id":"64bea31d81caff7f184ad4a0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64bea31d81caff7f184ad4a0/hq4f77BRNoiY92musz9kq.jpeg","isPro":false,"fullname":"Quan","user":"wq2012","type":"user"},{"_id":"62611fcabbcbd1c34f1615f6","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62611fcabbcbd1c34f1615f6/tXx0cbsnQM03EinbKMY0x.jpeg","isPro":false,"fullname":"Yoach Lacombe","user":"ylacombe","type":"user"},{"_id":"63b7e3814705f0ed5d7a0b00","avatarUrl":"/avatars/15cb3249c161b0266bbdce64b193a2a3.svg","isPro":false,"fullname":"Wang","user":"wsstriving","type":"user"}],"userCount":9,"collections":[{"slug":"diarizers-community/speaker-diarization-datasets-66261b8d571552066e003788","title":"Speaker Diarization Datasets","description":"A collection of speaker diarization datasets compatible with Diarizers.","gating":false,"lastUpdated":"2024-05-29T07:55:57.991Z","owner":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"items":[{"_id":"66261be022668df9ad5afbcd","position":0,"type":"dataset","author":"talkbank","downloads":882,"gated":"auto","id":"talkbank/callhome","lastModified":"2024-04-28T19:53:45.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":660,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":37,"isLikedByUser":false,"isBenchmark":false},{"_id":"66261bfe571552066e004b4c","position":2,"type":"dataset","author":"diarizers-community","downloads":82,"gated":false,"id":"diarizers-community/simsamu","lastModified":"2024-04-22T10:33:13.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":61,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":5,"isLikedByUser":false,"isBenchmark":false},{"_id":"66261c127fd76cbcf453fd67","position":3,"type":"dataset","author":"diarizers-community","downloads":1423,"gated":false,"id":"diarizers-community/ami","lastModified":"2024-04-22T10:34:25.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":212,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":1,"isLikedByUser":false,"isBenchmark":false},{"_id":"6656df83be448f6569d78c19","position":4,"type":"dataset","author":"talkbank","downloads":54,"gated":false,"id":"talkbank/sakura","lastModified":"2024-05-14T13:04:26.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":18,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":2,"isLikedByUser":false,"isBenchmark":false}],"position":1,"theme":"blue","private":false,"shareUrl":"https://hf.co/collections/diarizers-community/speaker-diarization-datasets","upvotes":6,"isUpvotedByUser":false},{"slug":"diarizers-community/models-66261d0f9277b825c807ff2a","title":"Models","description":"A collection of multilingual speaker segmentation model's fine-tuned using diarizers and compatible with pyannote. ","gating":false,"lastUpdated":"2024-04-24T15:50:54.962Z","owner":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"items":[{"_id":"66261d2182e322b92d91f619","position":0,"type":"model","author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":6750,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-eng","availableInferenceProviders":[],"lastModified":"2024-04-25T08:56:26.000Z","likes":5,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"_id":"66261d3bd78c254961354b12","position":1,"type":"model","author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":2,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-zho","availableInferenceProviders":[],"lastModified":"2024-04-25T08:56:48.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"_id":"66261d4947851d717eb3970f","position":2,"type":"model","author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":231,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-deu","availableInferenceProviders":[],"lastModified":"2024-04-25T08:55:48.000Z","likes":6,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"_id":"66261d580e31d65ecc5f15b6","position":3,"type":"model","author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":16,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-spa","availableInferenceProviders":[],"lastModified":"2024-04-25T08:57:09.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515}],"position":1,"theme":"green","private":false,"shareUrl":"https://hf.co/collections/diarizers-community/models","upvotes":1,"isUpvotedByUser":false},{"slug":"diarizers-community/diarizationlm-669dbae262b7eda846d75515","title":"DiarizationLM","description":"","gating":false,"lastUpdated":"2024-09-19T16:28:18.106Z","owner":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"items":[{"_id":"669dbaeb119595d21b91c049","position":0,"type":"paper","id":"2401.03506","title":"DiarizationLM: Speaker Diarization Post-Processing with Large Language\n Models","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2401.03506.png","upvotes":15,"publishedAt":"2024-01-07T14:54:57.000Z","isUpvotedByUser":false},{"_id":"669dbaf1f9a689cde92193de","position":1,"type":"model","author":"google","authorData":{"_id":"5e6aca39878b8b2bf9806447","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/WtA3YYitedOr9n02eHfJe.png","fullname":"Google","name":"google","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"enterprise","followerCount":45411,"isUserFollowing":false},"downloads":124,"gated":false,"id":"google/DiarizationLM-13b-Fisher-v1","availableInferenceProviders":[{"provider":"featherless-ai","modelStatus":"live","providerStatus":"live","providerId":"google/DiarizationLM-13b-Fisher-v1","task":"text-generation","adapterWeightsPath":"model-00001-of-00006.safetensors","isCheapestPricingOutput":false,"isFastestThroughput":false,"isModelAuthor":false}],"lastModified":"2024-08-11T21:51:33.000Z","likes":12,"pipeline_tag":"text-generation","private":false,"repoType":"model","isLikedByUser":false,"widgetOutputUrls":[],"numParameters":13015864320},{"_id":"669dbafe62b7eda846d75d88","position":2,"type":"model","author":"google","authorData":{"_id":"5e6aca39878b8b2bf9806447","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/WtA3YYitedOr9n02eHfJe.png","fullname":"Google","name":"google","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"enterprise","followerCount":45411,"isUserFollowing":false},"downloads":54,"gated":false,"id":"google/DiarizationLM-8b-Fisher-v1","availableInferenceProviders":[],"lastModified":"2024-08-02T21:20:23.000Z","likes":3,"pipeline_tag":"text-generation","private":false,"repoType":"model","isLikedByUser":false,"widgetOutputUrls":[],"numParameters":8030261248},{"_id":"66ad4e778d2c325307230724","position":3,"type":"model","author":"google","authorData":{"_id":"5e6aca39878b8b2bf9806447","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/WtA3YYitedOr9n02eHfJe.png","fullname":"Google","name":"google","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"enterprise","followerCount":45411,"isUserFollowing":false},"downloads":570,"gated":false,"id":"google/DiarizationLM-8b-Fisher-v2","availableInferenceProviders":[],"lastModified":"2024-08-02T21:48:27.000Z","likes":33,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":8030261248}],"position":2,"theme":"blue","private":false,"shareUrl":"https://hf.co/collections/diarizers-community/diarizationlm","upvotes":0,"isUpvotedByUser":false}],"datasets":[{"author":"diarizers-community","downloads":40,"gated":false,"id":"diarizers-community/ami_for_diarizationlm","lastModified":"2024-07-17T15:46:50.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":170,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"diarizers-community","downloads":147,"gated":false,"id":"diarizers-community/ami_ihm_with_transcripts","lastModified":"2024-07-15T15:27:51.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":152,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"diarizers-community","downloads":510,"gated":false,"id":"diarizers-community/voxconverse","lastModified":"2024-05-31T15:27:07.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":448,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":7,"isLikedByUser":false,"isBenchmark":false},{"author":"diarizers-community","downloads":59,"gated":false,"id":"diarizers-community/synthetic-speaker-diarization-dataset","lastModified":"2024-05-29T14:30:17.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":1584,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":2,"isLikedByUser":false,"isBenchmark":false},{"author":"diarizers-community","downloads":1423,"gated":false,"id":"diarizers-community/ami","lastModified":"2024-04-22T10:34:25.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":212,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":1,"isLikedByUser":false,"isBenchmark":false},{"author":"diarizers-community","downloads":82,"gated":false,"id":"diarizers-community/simsamu","lastModified":"2024-04-22T10:33:13.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":61,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":5,"isLikedByUser":false,"isBenchmark":false}],"models":[{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":5240,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-jpn","availableInferenceProviders":[],"lastModified":"2024-04-25T08:58:01.000Z","likes":5,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":16,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-spa","availableInferenceProviders":[],"lastModified":"2024-04-25T08:57:09.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":2,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-zho","availableInferenceProviders":[],"lastModified":"2024-04-25T08:56:48.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":6750,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-eng","availableInferenceProviders":[],"lastModified":"2024-04-25T08:56:26.000Z","likes":5,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515},{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"downloads":231,"gated":false,"id":"diarizers-community/speaker-segmentation-fine-tuned-callhome-deu","availableInferenceProviders":[],"lastModified":"2024-04-25T08:55:48.000Z","likes":6,"private":false,"repoType":"model","isLikedByUser":false,"numParameters":1473515}],"paperPreviews":[],"spaces":[{"author":"diarizers-community","authorData":{"_id":"661d57d708dd378c815c91d0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65e5c224a0a2f649dab3dfe4/sxiWKaStwTEEKUMQjdP-E.png","fullname":"diarizers-community","name":"diarizers-community","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":80,"isUserFollowing":false},"colorFrom":"yellow","colorTo":"purple","createdAt":"2024-06-27T01:43:30.000Z","emoji":"💬","id":"diarizers-community/DiarizationLM-GGUF","lastModified":"2024-08-03T21:17:54.000Z","likes":6,"pinned":false,"private":false,"sdk":"gradio","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-basic","requested":"cpu-basic"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"diarizers-community-diarizationlm-gguf.hf.space","stage":"READY"}],"sha":"0d6cabe07e52319aae98e55bb1fd3eb2665ebb83"},"title":"DiarizationLM GGUF","isLikedByUser":false,"ai_short_description":"Generate speaker-labeled text from conversations","ai_category":"Text Generation","trendingScore":0,"tags":["gradio","region:us"],"featured":false}],"buckets":[],"numBuckets":0,"numDatasets":6,"numModels":5,"numSpaces":2,"lastOrgActivities":[],"acceptLanguages":["*"],"canReadRepos":false,"canReadSpaces":false,"blogPosts":[],"currentRepoPage":0,"filters":{},"paperView":false}">AI & ML interests
Speaker diarization
diarizers-community aims to promote speaker diarization on the Hugging Face hub. It contains:
- A collection of multilingual speaker diarization datasets that are compatible with the diarizers library. They have been processed using diarizers scripts.
The available datasets are the CallHome (Japanese, Chinese, German, Spanish, English), AMI Corpus (English), Vox-Converse (English) and Simsamu (French). We aim to add more datasets in the future to better support speaker diarization on the Hub.
- A collection of multilingual fine-tuned segmentation model baselines compatible with pyannote.
Each model has been fine-tuned on a specific Callhome language subset. They achieve better performances on multilingual data compared to pyannote's pre-trained segmentation-3.0 model (see benchmark for more details on model performance).
Together with diarizers-community, we release:
diarizers, a library for fine-tuning pyannote speaker diarization models using the Hugging Face ecosystem.
A google colab notebook, with a step-by-step guide on how to use diarizers.
Benchmark
| Callhome test dataset | Model | DER | False alarm | Missed detection | Confusion |
|---|---|---|---|---|---|
| Japanese | Pretrained | 25.44 | 2.30 | 17.45 | 5.69 |
| Fine-tuned | 18.23 | 6.31 | 6.91 | 5.01 | |
| Spanish | Pretrained | 33.44 | 2.59 | 25.19 | 5.66 |
| Fine-tuned | 25.72 | 6.87 | 12.73 | 6.12 | |
| English | Pretrained | 22.16 | 6.29 | 10.97 | 4.90 |
| Fine-tuned | 18.40 | 7.10 | 6.98 | 4.32 | |
| German | Pretrained | 21.90 | 3.10 | 14.25 | 4.55 |
| Fine-tuned | 16.75 | 5.00 | 7.75 | 4.00 | |
| Chinese | Pretrained | 19.73 | 4.81 | 9.82 | 5.11 |
| Fine-tuned | 15.95 | 5.04 | 7.24 | 3.68 |
Results are in %. They have been obtained using the test script from diarizers.
-
diarizers-community/speaker-segmentation-fine-tuned-callhome-eng
1.47M • Updated • 6.75k • 5 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-zho
1.47M • Updated • 2 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-deu
1.47M • Updated • 231 • 6 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-spa
1.47M • Updated • 16
-
diarizers-community/speaker-segmentation-fine-tuned-callhome-eng
1.47M • Updated • 6.75k • 5 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-zho
1.47M • Updated • 2 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-deu
1.47M • Updated • 231 • 6 -
diarizers-community/speaker-segmentation-fine-tuned-callhome-spa
1.47M • Updated • 16