From add7bd8ef86ec21cd1327eb45bcb739aa54f7db8 Mon Sep 17 00:00:00 2001 From: yum Date: Thu, 5 Oct 2023 18:22:55 -0700 Subject: Transcripts preceding long pauses now drop When hot-miking into the built-in chatbox, there are sometimes long pauses in conversation. After these pauses, it's undesirable to show the transcript generate before the pause. This feature makes it so that those transcripts can be dropped. Also: * Limit number of segments sent to browser source to 10. Allow this to grow up to 10 segments before dropping the first 5 segments. * Silence warnings generated by `install_in_venv`, used by e.g. translation codepath. * Enable audio normalization to improve accuracy when speaking softly, at the cost of some accuracy when speaking normally. Credit: user endo0269 on Discord suggested this feature. --- Scripts/transcribe_pipeline.py | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'Scripts/transcribe_pipeline.py') diff --git a/Scripts/transcribe_pipeline.py b/Scripts/transcribe_pipeline.py index 3f48b08..5914afc 100644 --- a/Scripts/transcribe_pipeline.py +++ b/Scripts/transcribe_pipeline.py @@ -5,15 +5,22 @@ class TranscriptCommit: def __init__(self, delta: str, preview: str, - latency_s: int = None, + latency_s: float = None, thresh_at_commit: int = None, - audio: bytes = None): + audio: bytes = None, + duration_s: float = None, + start_ts: float = None): self.delta = delta self.preview = preview self.latency_s = latency_s self.thresh_at_commit = thresh_at_commit self.audio = audio + # Time at which the commit is generated self.ts = time.time() + # Time corresponding to the start of the segment + self.start_ts = start_ts + # The duration of the audio segment, in seconds. + self.duration_s = duration_s class StreamingPlugin: -- cgit v1.2.3