Projects - Article Read-aloud-er

Overview

This project provides a Web (or mobile) service to read documents/articles aloud.

We use

Features

• Docs imported from

• Continuous Reading with looping among the

Similar alternatives;

• OneNote Immersive Reader - https://support.microsoft.com/ja-jp/topic/onenote-のイマーシブ-リーダー機能を使用する-10712138-b4ed-4513-958d-d9a1b3038170

Notes

• Learning about APIs

  • OneNote Investigations - OneNote

  • TTS

• Trying Immersive Reader SDK - https://learn.microsoft.com/en-us/azure/applied-ai-services/immersive-reader/quickstarts/client-libraries?pivots=programming-language-nodejs / https://learn.microsoft.com/en-us/azure/applied-ai-services/immersive-reader/reference

  • need to follow https://learn.microsoft.com/en-us/azure/applied-ai-services/immersive-reader/how-to-create-immersive-reader to create tokens

  • NOTE: https://learn.microsoft.com/en-us/azure/applied-ai-services/immersive-reader/reference#readaloudoptions this table has autoPlay (capital P) but actual working property name is autoplay.

  • Can we “hack” by messages..?

• Trying OneNote APIs

Considerations

• over Immersive Reader?

  • continuous/ambient reading i.e. proceeding the the next document automatically

• https://learn.microsoft.com/en-us/azure/applied-ai-services/immersive-reader/

  • Immersive Reader itself is also an option~

• Text modifications for PDF

  • We might workaround this for Immersive Reader by converting PDF to html/text first though.

  • less

    • TOOD: Investigate more

  • pdftohtml

    • Does not process “column layout” well (without proper metadata embedded?)

    • But did better than OneNote’s OCR in some cases.

Reference

• Document Storage APIs

  • https://learn.microsoft.com/ja-jp/graph/integrate-with-onenote

  • https://dev.evernote.com/

• Text-to-Speech APIs/Libs

  • https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/rest-text-to-speech?tabs=streaming

    • https://azure.microsoft.com/ja-jp/products/cognitive-services/text-to-speech

  • https://cloud.google.com/text-to-speech?hl=ja

  • https://medevel.com/14-os-text-to-speech/

• PDF Text Extraction

  • https://pdf.wondershare.com/pdf-knowledge/pdf-to-html-linux.html

    • pdftohtml

• VOIVEVOX

  • https://github.com/VOICEVOX/voicevox_engine/

  • https://vtubermochio.wixsite.com/mochizora/もち子さんとは-設定資料 / https://voicevox.hiroshiba.jp/dormitory/mochikosan/

    • https://voicevox.hiroshiba.jp/dormitory/

• OpenTTS

  • https://github.com/synesthesiam/opentts

    • I like

      • marytts: cmu-slt-hsmm [female] (en_us)?

      • larynx:ljspeech-glow_tts

      • larynx:glados-glow_tts