While recent text to speech (TTS) models perform very well in synthesizing reading-style (e.g., audiobook) speech, it is still challenging to synthesize spontaneous-style speech (e.g., podcast or conversation), mainly because of two reasons: 1) the lack of training data for spontaneous speech; 2) the difficulty in modeling the filled pauses (um and uh) and diverse rhythms in spontaneous speech. In this paper, we develop AdaSpeech 3, an adaptive TTS system that fine-tunes a well-trained reading-style TTS model for spontaneous-style speech. Specifically, 1) to insert filled pauses (FP) in the text sequence appropriately, we introduce an FP predictor to the TTS model; 2) to model the varying rhythms, we introduce a duration predictor based on mixture of experts (MoE), which contains three experts responsible for the generation of fast, medium and slow speech respectively, and fine-tune it as well as the pitch predictor for rhythm adaptation; 3) to adapt to other speaker timbre, we fine-tune some parameters in the decoder with few speech data. To address the challenge of lack of training data, we mine a spontaneous speech dataset to support our research this work and facilitate future research on spontaneous TTS. Experiments show that AdaSpeech 3 synthesizes speech with natural FP and rhythms in spontaneous styles, and achieves much better MOS and SMOS scores than previous adaptive TTS systems.<\/p>\n","protected":false},"excerpt":{"rendered":"
While recent text to speech (TTS) models perform very well in synthesizing reading-style (e.g., audiobook) speech, it is still challenging to synthesize spontaneous-style speech (e.g., podcast or conversation), mainly because of two reasons: 1) the lack of training data for spontaneous speech; 2) the difficulty in modeling the filled pauses (um and uh) and diverse […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13545],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-field-of-study":[257728,246691,249049,256420,256207,256435,247753,249400,258898,258895],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-767887","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-human-language-technologies","msr-locale-en_us","msr-field-of-study-adaptation-computer-science","msr-field-of-study-computer-science","msr-field-of-study-conversation","msr-field-of-study-duration-music","msr-field-of-study-natural-music","msr-field-of-study-rhythm","msr-field-of-study-speech-recognition","msr-field-of-study-speech-synthesis","msr-field-of-study-style-sociolinguistics","msr-field-of-study-timbre"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2021-7-5","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/arxiv.org\/abs\/2107.02530","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Yuzi Yan","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Xu Tan","user_id":37116,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Xu Tan"},{"type":"text","value":"Bohan Li","user_id":0,"rest_url":false},{"type":"text","value":"Guangyan Zhang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Tao Qin","user_id":33871,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Tao Qin"},{"type":"text","value":"Sheng Zhao","user_id":0,"rest_url":false},{"type":"text","value":"Yuan Shen","user_id":0,"rest_url":false},{"type":"text","value":"Wei-Qiang Zhang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Tie-Yan Liu","user_id":34431,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Tie-Yan Liu"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[766678],"msr_group":[],"msr_project":[],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/767887"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/767887\/revisions"}],"predecessor-version":[{"id":767893,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/767887\/revisions\/767893"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=767887"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=767887"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=767887"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=767887"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=767887"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=767887"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=767887"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=767887"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=767887"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=767887"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=767887"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=767887"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=767887"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=767887"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=767887"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}