November 2, 2019 Conference paper Open Access

Di Gangi, Matti; Enyedi, Robert; Brusadin, Alessandra; Federico, Marcello

JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/6f302ec4-375b-4d4c-8518-8b5d42d770c1/IWSLT2019_paper_3.pdf"
      }, 
      "checksum": "md5:3d5686aed1fb4127847cffda9e8a5f5d", 
      "bucket": "6f302ec4-375b-4d4c-8518-8b5d42d770c1", 
      "key": "IWSLT2019_paper_3.pdf", 
      "type": "pdf", 
      "size": 139777
    }
  ], 
  "owners": [
    50447
  ], 
  "doi": "10.5281/zenodo.3524947", 
  "stats": {
    "version_unique_downloads": 143.0, 
    "unique_views": 666.0, 
    "views": 696.0, 
    "version_views": 696.0, 
    "unique_downloads": 143.0, 
    "version_unique_views": 666.0, 
    "volume": 22504097.0, 
    "version_downloads": 161.0, 
    "downloads": 161.0, 
    "version_volume": 22504097.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.3524947", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.3524946", 
    "bucket": "https://zenodo.org/api/files/6f302ec4-375b-4d4c-8518-8b5d42d770c1", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.3524946.svg", 
    "html": "https://zenodo.org/record/3524947", 
    "latest_html": "https://zenodo.org/record/3524947", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.3524947.svg", 
    "latest": "https://zenodo.org/api/records/3524947"
  }, 
  "conceptdoi": "10.5281/zenodo.3524946", 
  "created": "2019-11-01T11:36:23.016627+00:00", 
  "updated": "2020-01-20T17:34:02.488503+00:00", 
  "conceptrecid": "3524946", 
  "revision": 2, 
  "id": 3524947, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.3524947", 
    "description": "<p>Neural machine translation models have shown to achieve high quality when trained and fed with well structured and punctuated input&nbsp;texts. Unfortunately, the latter condition is not met in spoken language translation, where the input is generated by an automatic speech&nbsp;recognition (ASR) system. In this paper, we study how to adapt a strong NMT system to make it robust to typical ASR errors. As in our application scenarios transcripts might be post-edited by human experts, we propose adaptation strategies to train a single system that can translate either clean or noisy input with no supervision on the input type. Our experimental results on a public speech translation data set show that adapting a model on a significant amount of parallel data including ASR transcripts is beneficial with test data of the same type, but produces a small degradation when translating clean text. Adapting on both clean and noisy variants of the same data leads to the best results on both input types.</p>", 
    "language": "eng", 
    "title": "Robust Neural Machine Translation for Clean and Noisy Speech Transcripts", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3524946"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "3524947"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "iwslt2019"
      }
    ], 
    "publication_date": "2019-11-02", 
    "creators": [
      {
        "affiliation": "Fondazione Bruno Kessler, Trento, Italy & University of Trento, Italy", 
        "name": "Di Gangi, Matti"
      }, 
      {
        "affiliation": "Amazon AI, East Palo Alto, USA", 
        "name": "Enyedi, Robert"
      }, 
      {
        "affiliation": "Amazon AI, East Palo Alto, USA", 
        "name": "Brusadin, Alessandra"
      }, 
      {
        "affiliation": "Amazon AI, East Palo Alto, USA", 
        "name": "Federico, Marcello"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }, 
    "related_identifiers": [
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.3524946", 
        "relation": "isVersionOf"
      }
    ]
  }
}