*Contact emails*
[email protected], [email protected]

*Explainer*
github.com/WebAudio/web-speech-api/blob/main/explainers/speech-recognition-result-timestamps.md
 
*Specification*
github.com/WebAudio/web-speech-api/pull/192 
*Summary*
The Web Speech API currently does not allow developers to associate the 
transcribed text with specific segments of the audio source, making it 
difficult to map the generated text to the timeline or determine on-device 
transcription latency. We propose adding audioStartTime and audioEndTime (
DOMHighResTimeStamp?) attributes on the SpeechRecognitionResult interface.

*Blink component*
Blink>Speech 
<https://issues.chromium.org/issues?q=customfield1222907:%22Blink%3ESpeech%22>

*Web Feature ID*
speech-recognition <https://webstatus.dev/features/speech-recognition>

*Motivation*

Web applications are increasingly adopting on-device Automatic Speech 
Recognition (ASR) via the Web Speech API (SpeechRecognition) to enhance 
user privacy and lower backend processing costs. However, the WebSpeech API 
operates as a "black box" regarding local processing delays:


   1. 
   
   Lack of Latency Tracking & Backend Recovery: When client devices with 
   constrained CPU/GPU resources struggle to process audio in real time, 
   significant transcription lag accumulates between when audio was spoken and 
   when recognition results are emitted. Web applications have no web-exposed 
   timing metrics to programmatically detect this lag or trigger an optional 
   failover to a high-quality cloud ASR backend provider, resulting in 
   degraded user experiences (e.g., stale or delayed captions in live video 
   conferencing tools).
   2. 
   
   Timeline Association: Developers cannot readily map recognized 
   transcript text to specific segments of the source audio timeline, creating 
   challenges for automated subtitle cue alignment and media editing tools.
   

*Initial public proposal*
https://github.com/WebAudio/web-speech-api/issues/191 
*Requires code in //chrome?*
True (initial integration with Speech Recognition service and SODA backend; 
later with ODML/LLM backend).

*Tracking bug*
crbug.com/528037568


*Estimated milestones*Target DevTrial: M153

*Link to entry on the Chrome Platform Status*
https://chromestatus.com/feature/5811907077472256

This intent message was generated by Chrome Platform Status 
<https://chromestatus.com/>.

-- 
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/23190a02-f4af-476d-8721-92f438b12aecn%40chromium.org.

Reply via email to