Changes

              
    

          
          
        
        

            
f
1
{
f
1
{
            
2
  "access_rights": "",
2
  "access_rights": "",
            
3
  "author": "Alexander H. Liu",
3
  "author": "Alexander H. Liu",
            
4
  "author_email": "",
4
  "author_email": "",
            
5
  "citation": [],
5
  "citation": [],
            
6
  "creator_user_id": "17755db4-395a-4b3b-ac09-e8e3484ca700",
6
  "creator_user_id": "17755db4-395a-4b3b-ac09-e8e3484ca700",
            
7
  "defined_in": "https://doi.org/10.48550/arXiv.2310.16338",
7
  "defined_in": "https://doi.org/10.48550/arXiv.2310.16338",
            
8
  "doi": "10.57702/74wogvts",
8
  "doi": "10.57702/74wogvts",
            
n
9
  "doi_date_published": null,
n
9
  "doi_date_published": "2024-12-03",
            
10
  "doi_publisher": "TIB",
10
  "doi_publisher": "TIB",
            
n
11
  "doi_status": false,
n
11
  "doi_status": true,
            
12
  "domain": "https://service.tib.eu/ldmservice",
12
  "domain": "https://service.tib.eu/ldmservice",
            
13
  "extra_authors": [
13
  "extra_authors": [
            
14
    {
14
    {
            
15
      "extra_author": "Matt Le",
15
      "extra_author": "Matt Le",
            
16
      "orcid": ""
16
      "orcid": ""
            
17
    },
17
    },
            
18
    {
18
    {
            
19
      "extra_author": "Apoorv Vyas",
19
      "extra_author": "Apoorv Vyas",
            
20
      "orcid": ""
20
      "orcid": ""
            
21
    },
21
    },
            
22
    {
22
    {
            
23
      "extra_author": "Bowen Shi",
23
      "extra_author": "Bowen Shi",
            
24
      "orcid": ""
24
      "orcid": ""
            
25
    },
25
    },
            
26
    {
26
    {
            
27
      "extra_author": "Andros Tjandra",
27
      "extra_author": "Andros Tjandra",
            
28
      "orcid": ""
28
      "orcid": ""
            
29
    },
29
    },
            
30
    {
30
    {
            
31
      "extra_author": "Wei-Ning Hsu",
31
      "extra_author": "Wei-Ning Hsu",
            
32
      "orcid": ""
32
      "orcid": ""
            
33
    }
33
    }
            
34
  ],
34
  ],
            
35
  "groups": [
35
  "groups": [
            
36
    {
36
    {
            
37
      "description": "",
37
      "description": "",
            
38
      "display_name": "Speech Enhancement",
38
      "display_name": "Speech Enhancement",
            
39
      "id": "3ebe1a8a-9a26-403d-9e47-c18cf0819c13",
39
      "id": "3ebe1a8a-9a26-403d-9e47-c18cf0819c13",
            
40
      "image_display_url": "",
40
      "image_display_url": "",
            
41
      "name": "speech-enhancement",
41
      "name": "speech-enhancement",
            
42
      "title": "Speech Enhancement"
42
      "title": "Speech Enhancement"
            
43
    },
43
    },
            
44
    {
44
    {
            
45
      "description": "",
45
      "description": "",
            
46
      "display_name": "Speech Separation",
46
      "display_name": "Speech Separation",
            
47
      "id": "898035a0-64b4-4a27-a600-539b39909d29",
47
      "id": "898035a0-64b4-4a27-a600-539b39909d29",
            
48
      "image_display_url": "",
48
      "image_display_url": "",
            
49
      "name": "speech-separation",
49
      "name": "speech-separation",
            
50
      "title": "Speech Separation"
50
      "title": "Speech Separation"
            
51
    },
51
    },
            
52
    {
52
    {
            
53
      "description": "",
53
      "description": "",
            
54
      "display_name": "Speech Synthesis",
54
      "display_name": "Speech Synthesis",
            
55
      "id": "3b8f8411-cbd1-410f-92f8-5ecbc4095d46",
55
      "id": "3b8f8411-cbd1-410f-92f8-5ecbc4095d46",
            
56
      "image_display_url": "",
56
      "image_display_url": "",
            
57
      "name": "speech-synthesis",
57
      "name": "speech-synthesis",
            
58
      "title": "Speech Synthesis"
58
      "title": "Speech Synthesis"
            
59
    }
59
    }
            
60
  ],
60
  ],
            
61
  "id": "9b28f7ed-2d16-495c-a740-15b8d6ffa9db",
61
  "id": "9b28f7ed-2d16-495c-a740-15b8d6ffa9db",
            
62
  "isopen": false,
62
  "isopen": false,
            
63
  "landing_page": "https://voicebox.metademolab.com/speechflow.html",
63
  "landing_page": "https://voicebox.metademolab.com/speechflow.html",
            
64
  "license_title": null,
64
  "license_title": null,
            
65
  "link_orkg": "",
65
  "link_orkg": "",
            
66
  "metadata_created": "2024-12-03T10:19:29.895579",
66
  "metadata_created": "2024-12-03T10:19:29.895579",
            
n
67
  "metadata_modified": "2024-12-03T10:19:29.895585",
n
67
  "metadata_modified": "2024-12-03T10:19:30.244844",
            
68
  "name": "generative-pre-training-for-speech",
68
  "name": "generative-pre-training-for-speech",
            
69
  "notes": "Generative models have gained more and more attention in 
69
  "notes": "Generative models have gained more and more attention in 
            
70
recent years for their remarkable success in tasks that required 
70
recent years for their remarkable success in tasks that required 
            
71
estimating and sampling data distribution to generate high-fidelity 
71
estimating and sampling data distribution to generate high-fidelity 
            
72
synthetic data. In speech, text-to-speech synthesis and neural vocoder 
72
synthetic data. In speech, text-to-speech synthesis and neural vocoder 
            
73
are good examples where generative models have shined. While 
73
are good examples where generative models have shined. While 
            
74
generative models have been applied to different applications in 
74
generative models have been applied to different applications in 
            
75
speech, there exists no general-purpose generative model that models 
75
speech, there exists no general-purpose generative model that models 
            
76
speech directly. In this work, we take a step toward this direction by 
76
speech directly. In this work, we take a step toward this direction by 
            
77
showing a single pre-trained generative model can be adapted to 
77
showing a single pre-trained generative model can be adapted to 
            
78
different downstream tasks with strong performance.",
78
different downstream tasks with strong performance.",
            
n
79
  "num_resources": 0,
n
79
  "num_resources": 1,
            
80
  "num_tags": 4,
80
  "num_tags": 4,
            
81
  "organization": {
81
  "organization": {
            
82
    "approval_status": "approved",
82
    "approval_status": "approved",
            
83
    "created": "2024-11-25T12:11:38.292601",
83
    "created": "2024-11-25T12:11:38.292601",
            
84
    "description": "",
84
    "description": "",
            
85
    "id": "079d46db-32df-4b48-91f3-0a8bc8f69559",
85
    "id": "079d46db-32df-4b48-91f3-0a8bc8f69559",
            
86
    "image_url": "",
86
    "image_url": "",
            
87
    "is_organization": true,
87
    "is_organization": true,
            
88
    "name": "no-organization",
88
    "name": "no-organization",
            
89
    "state": "active",
89
    "state": "active",
            
90
    "title": "No Organization",
90
    "title": "No Organization",
            
91
    "type": "organization"
91
    "type": "organization"
            
92
  },
92
  },
            
93
  "owner_org": "079d46db-32df-4b48-91f3-0a8bc8f69559",
93
  "owner_org": "079d46db-32df-4b48-91f3-0a8bc8f69559",
            
94
  "private": false,
94
  "private": false,
            
95
  "relationships_as_object": [],
95
  "relationships_as_object": [],
            
96
  "relationships_as_subject": [],
96
  "relationships_as_subject": [],
            
t
97
  "resources": [],
t
97
  "resources": [
            
98
    {
            
99
      "cache_last_updated": null,
            
100
      "cache_url": null,
            
101
      "created": "2024-12-03T11:11:17",
            
102
      "data": [
            
103
        "dcterms:title",
            
104
        "dcterms:accessRights",
            
105
        "dcterms:creator",
            
106
        "dcterms:description",
            
107
        "dcterms:issued",
            
108
        "dcterms:language",
            
109
        "dcterms:identifier",
            
110
        "dcat:theme",
            
111
        "dcterms:type",
            
112
        "dcat:keyword",
            
113
        "dcat:landingPage",
            
114
        "dcterms:hasVersion",
            
115
        "dcterms:format",
            
116
        "mls:task",
            
117
        "datacite:isDescribedBy"
            
118
      ],
            
119
      "description": "The json representation of the dataset with its 
            
120
distributions based on DCAT.",
            
121
      "format": "JSON",
            
122
      "hash": "",
            
123
      "id": "1747e102-ae4d-41ec-bbba-ab1a9cf0513c",
            
124
      "last_modified": "2024-12-03T10:19:30.236763",
            
125
      "metadata_modified": "2024-12-03T10:19:30.247833",
            
126
      "mimetype": "application/json",
            
127
      "mimetype_inner": null,
            
128
      "name": "Original Metadata",
            
129
      "package_id": "9b28f7ed-2d16-495c-a740-15b8d6ffa9db",
            
130
      "position": 0,
            
131
      "resource_type": null,
            
132
      "size": 1418,
            
133
      "state": "active",
            
134
      "url": 
            
135
resource/1747e102-ae4d-41ec-bbba-ab1a9cf0513c/download/metadata.json",
            
136
      "url_type": "upload"
            
137
    }
            
138
  ],
            
98
  "services_used_list": "",
139
  "services_used_list": "",
            
99
  "state": "active",
140
  "state": "active",
            
100
  "tags": [
141
  "tags": [
            
101
    {
142
    {
            
102
      "display_name": "Generative Models",
143
      "display_name": "Generative Models",
            
103
      "id": "dafd9c1f-394f-4d88-9c07-5801683e057a",
144
      "id": "dafd9c1f-394f-4d88-9c07-5801683e057a",
            
104
      "name": "Generative Models",
145
      "name": "Generative Models",
            
105
      "state": "active",
146
      "state": "active",
            
106
      "vocabulary_id": null
147
      "vocabulary_id": null
            
107
    },
148
    },
            
108
    {
149
    {
            
109
      "display_name": "Speech Enhancement",
150
      "display_name": "Speech Enhancement",
            
110
      "id": "04c547c7-e18e-475d-8ddc-39327b20ffb2",
151
      "id": "04c547c7-e18e-475d-8ddc-39327b20ffb2",
            
111
      "name": "Speech Enhancement",
152
      "name": "Speech Enhancement",
            
112
      "state": "active",
153
      "state": "active",
            
113
      "vocabulary_id": null
154
      "vocabulary_id": null
            
114
    },
155
    },
            
115
    {
156
    {
            
116
      "display_name": "Speech Separation",
157
      "display_name": "Speech Separation",
            
117
      "id": "7ac466f4-12eb-4865-af7f-d204f9ccb1ba",
158
      "id": "7ac466f4-12eb-4865-af7f-d204f9ccb1ba",
            
118
      "name": "Speech Separation",
159
      "name": "Speech Separation",
            
119
      "state": "active",
160
      "state": "active",
            
120
      "vocabulary_id": null
161
      "vocabulary_id": null
            
121
    },
162
    },
            
122
    {
163
    {
            
123
      "display_name": "Speech Synthesis",
164
      "display_name": "Speech Synthesis",
            
124
      "id": "b648c916-18f5-4a18-9e4e-22bd013ac34b",
165
      "id": "b648c916-18f5-4a18-9e4e-22bd013ac34b",
            
125
      "name": "Speech Synthesis",
166
      "name": "Speech Synthesis",
            
126
      "state": "active",
167
      "state": "active",
            
127
      "vocabulary_id": null
168
      "vocabulary_id": null
            
128
    }
169
    }
            
129
  ],
170
  ],
            
130
  "title": "Generative Pre-Training for Speech",
171
  "title": "Generative Pre-Training for Speech",
            
131
  "type": "dataset",
172
  "type": "dataset",
            
132
  "version": ""
173
  "version": ""
            
133
}
174
}