{ "access": { "embargo": { "active": false, "reason": null }, "files": "public", "record": "public", "status": "open" }, "created": "2017-06-27T13:24:51.885097+00:00", "custom_fields": {}, "deletion_status": { "is_deleted": false, "status": "P" }, "files": { "count": 1, "enabled": true, "entries": { "zrsc17-track1-IRIT-UPS.tar.gz": { "checksum": "md5:4996b08277b02c0f8c48b855be7a651a", "ext": "gz", "id": "4a4c3276-fc01-48dc-bfe8-6984339fc60f", "key": "zrsc17-track1-IRIT-UPS.tar.gz", "metadata": null, "mimetype": "application/gzip", "size": 10748318208 } }, "order": [], "total_bytes": 10748318208 }, "id": "819892", "is_draft": false, "is_published": true, "links": { "access": "https://zenodo.org/api/records/819892/access", "access_links": "https://zenodo.org/api/records/819892/access/links", "access_request": "https://zenodo.org/api/records/819892/access/request", "access_users": "https://zenodo.org/api/records/819892/access/users", "archive": "https://zenodo.org/api/records/819892/files-archive", "archive_media": "https://zenodo.org/api/records/819892/media-files-archive", "communities": "https://zenodo.org/api/records/819892/communities", "communities-suggestions": "https://zenodo.org/api/records/819892/communities-suggestions", "doi": "https://doi.org/10.5281/zenodo.819892", "draft": "https://zenodo.org/api/records/819892/draft", "files": "https://zenodo.org/api/records/819892/files", "latest": "https://zenodo.org/api/records/819892/versions/latest", "latest_html": "https://zenodo.org/records/819892/latest", "media_files": "https://zenodo.org/api/records/819892/media-files", "parent": "https://zenodo.org/api/records/819891", "parent_doi": "https://zenodo.org/doi/10.5281/zenodo.819891", "parent_html": "https://zenodo.org/records/819891", "requests": "https://zenodo.org/api/records/819892/requests", "reserve_doi": "https://zenodo.org/api/records/819892/draft/pids/doi", "self": "https://zenodo.org/api/records/819892", "self_doi": "https://zenodo.org/doi/10.5281/zenodo.819892", "self_html": "https://zenodo.org/records/819892", "self_iiif_manifest": "https://zenodo.org/api/iiif/record:819892/manifest", "self_iiif_sequence": "https://zenodo.org/api/iiif/record:819892/sequence/default", "versions": "https://zenodo.org/api/records/819892/versions" }, "media_files": { "count": 0, "enabled": false, "entries": {}, "order": [], "total_bytes": 0 }, "metadata": { "creators": [ { "affiliations": [ { "name": "Universit\u00e9 de Toulouse ; IRIT" } ], "person_or_org": { "family_name": "Pellegrini", "given_name": "Thomas", "name": "Pellegrini, Thomas", "type": "personal" } }, { "affiliations": [ { "name": "Universit\u00e9 de Toulouse ; IRIT" } ], "person_or_org": { "family_name": "Manenti", "given_name": "C\u00e9line", "name": "Manenti, C\u00e9line", "type": "personal" } }, { "affiliations": [ { "name": "Universit\u00e9 de Toulouse ; IRIT" } ], "person_or_org": { "family_name": "Pinquier", "given_name": "Julien", "name": "Pinquier, Julien", "type": "personal" } } ], "description": "
We run a k-means (k=100 clusters, L2 norm) on the baseline MFCCs (static, first and second derivative, 39-d) that are ZCA whitened on a per-file basis. The feature representations correspond to the distances between the data points and the cluster centroids.
", "publication_date": "2017-06-27", "publisher": "Zenodo", "resource_type": { "id": "software", "title": { "de": "Software", "en": "Software" } }, "rights": [ { "description": { "en": "The Creative Commons Attribution license allows re-distribution and re-use of a licensed work on the condition that the creator is appropriately credited." }, "icon": "cc-by-icon", "id": "cc-by-4.0", "props": { "scheme": "spdx", "url": "https://creativecommons.org/licenses/by/4.0/legalcode" }, "title": { "en": "Creative Commons Attribution 4.0 International" } } ], "subjects": [ { "subject": "zerospeech2017" } ], "title": "UPS-IRIT system #1 for track1" }, "parent": { "access": { "owned_by": { "user": 33255 } }, "communities": { "default": "9d0f3d04-8a55-4a2c-88d7-4d2296b36eb0", "entries": [ { "access": { "member_policy": "open", "members_visibility": "public", "record_policy": "open", "review_policy": "open", "visibility": "public" }, "children": { "allow": false }, "created": "2017-05-30T10:13:47.211546+00:00", "custom_fields": {}, "deletion_status": { "is_deleted": false, "status": "P" }, "id": "9d0f3d04-8a55-4a2c-88d7-4d2296b36eb0", "links": {}, "metadata": { "curation_policy": "Contributions are accepted from participants of the challenge.
\r\n", "page": "The ZeroSpeech challenge targets the unsupervised discovery of linguistic units from raw speech in an unknown language. As in the 2015 edition, it concerns two core components: the discovery of subword units (Track 1) and the discovery of word units (Track 2), respectively. For both tracks, the same evaluation metrics as in the former edition are used.
\r\n\r\nMore information can be found at www.zerospeech.com
\r\n\r\nThe ultimate goal of the Zero Resource Speech Challenge is to construct a system that learn an end-to-end Spoken Dialog (SD) system, in an unknown language, from scratch, using only information available to a language learning infant. "Zero resource" refers to zero linguistic expertise (e.g., orthographic/linguistic transcriptions), not zero information besides audio (visual, limited human feedback, etc). The fact that 4 year olds spontaneously learn language without supervision from language experts show that this goal is theoretically reachable.
\r\n\r\nThe Zero Resource speech challenge addresses a fundamental scientific question (how can a system autonomously acquire language?) which is interesting in its own right, but has also three main practical benefits:
\r\n\r\nThe Zero Resource Challenge series is constructed to progress incrementally towards this goal, by proposing achievable but progressively harder objectives, building and open sourcing the core technological components that are needed for an autonomous SD system along the way.
\r\n\r\nWeakly/Un- supervised learning is tricky to evaluate. We use two kinds of evaluation principles: (1). Unit testing: Each core component is evaluated by a specific set of metrics, largely inspired by psychometrics and linguistics. These tests do not guarantee that an entire system will work well, but they are useful to check and debug the systems. (2). Application testing. As the challenge progress in aggregating more components, useful applications will be possible to construct (e.g. keyword search, document classification, image retrieval from speech, speech to speech translation, etc), making it possible to use more standard evaluation techniques.
", "title": "ZeroSpeech2017" }, "revision_id": 0, "slug": "zerospeech2017", "updated": "2017-05-30T11:56:53.909545+00:00" } ], "ids": [ "9d0f3d04-8a55-4a2c-88d7-4d2296b36eb0" ] }, "id": "819891", "pids": { "doi": { "client": "datacite", "identifier": "10.5281/zenodo.819891", "provider": "datacite" } } }, "pids": { "doi": { "client": "datacite", "identifier": "10.5281/zenodo.819892", "provider": "datacite" }, "oai": { "identifier": "oai:zenodo.org:819892", "provider": "oai" } }, "revision_id": 12, "stats": { "all_versions": { "data_volume": 139728136704.0, "downloads": 13, "unique_downloads": 11, "unique_views": 123, "views": 142 }, "this_version": { "data_volume": 139728136704.0, "downloads": 13, "unique_downloads": 11, "unique_views": 123, "views": 142 } }, "status": "published", "updated": "2020-01-25T07:26:42.826241+00:00", "versions": { "index": 1, "is_latest": true } }