TTS/STT Use cases

[guest271314]

Text-To-Speech (TTS)

While the Web Speech API does provide basic capability to communicate with the local speech dispatcher socket speechd some functionality is not available, including

• SSML input
• File input
• File output

Speech-To-Text (STT)

Currently only Chrome, Chromium browsers implement STT. That functionality is based on recording the user voice (or other audio output) locally and sending that recording to a remote server, therefore, also some functionality is not available, including

• Local STT processing
• File input
• File output

Current workarounds

• https://github.com/guest271314/SpeechSynthesisRecorder (record audio output of speechSynthesis.speak() to get a file output)
• https://github.com/guest271314/SpeechSynthesisSSMLParser (parse SSML before executing speechSynthesis.speak() where espeak and espeak-ng accept SSML input directly)

TL;DR

• How to programmatically send a unix socket command to a system server autospawned by browser or convert JavaScript to C++ souce code for Chromium?
• https://bugzilla.mozilla.org/show_bug.cgi?id=1425523
• https://bugs.chromium.org/p/chromium/issues/detail?id=795371
• How is a complete SSML document expected to be parsed when set once at .text property of SpeechSynthesisUtterance instance? WICG/speech-api#10
• SpeechRecognition MUST be granted permission to record and send users voice to a remote web service WICG/speech-api#56
• Pipsqueak Engine as service for Web Speech API SpeechRecognition at desktop? mozilla/DeepSpeech#2401

Direct communication with one or more local TTS/STT applications and control of local audio input and output devices over HTTPS/WSS are the use cases descibed in this issue.

Can file a PR for this if necessary.