- 1、本文档共19页,可阅读全部内容。
- 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
- 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载。
- 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
RetrievalofMandarinChinese
Retrieval of Mandarin Chinese
Broadcast News via Speech
Berlin Chen
Graduate Institute of Computer Science Information Engineering
National Taiwan Normal University
Outline
2
What is spoken document retrieval (SDR)
Related work and current progress
Mandarin SDR research at NTU and Sinica
A prototype system for Mandarin SDR
Conclusions and future work
Spoken Document Retrieval
3
Main Task
Automatically indexing a collection of spoken documents with
speech recognition techniques
Retrieving relevant documents in response to a text/speech
query
Why Interesting?
Huge quantities of spoken/audio documents are becoming
available, such as broadcast news, digital library archives, etc.
Speech is currently a difficult media to browse and search
through
Possible use of speech as a new data type for retrieval in
the future
Speech provides a better user interface in wireless
environments
Scenario for Information Retrieval
4
speech Text-to-
information Speech text
Synthesis information
Public Services/
Information/
Knowledge
h
文档评论(0)