Generate timecoded SRT subtitles from local video or audio files. Use when a user wants a local low-cost subtitle workflow, asks to transcribe local media in...
18 次安装1 星标
设计与多媒体
Subtitle Srt Export
试用Export a standard .srt subtitle file from the MEASURED word timings in a job's input.json (written by elevenlabs-tts). Flattens the per-word timings into rea...
它能做什么
Export a standard .srt subtitle file from the MEASURED word timings in a job's input.json (written by elevenlabs-tts). Flattens the per-word timings into readable cues (max ~3.5s / ~10 words per cue, breaking on sentence punctuation) and writes <job>/subtitles.srt for upload to YouTube and other platforms. No API call. Re-runnable. Does not modify input.json.
技能文档
subtitle-srt-export
Turn the measured word-level timings into a ready-to-upload .srt file. No API
call, never mutates input.json, safe to re-run.
Why this skill exists
elevenlabs-tts writes accurate per-word timings into input.json's
subtitles[].words[]. Those word timings are perfect for an on-screen karaoke
render but are too granular for a subtitle file — a platform wants a handful of
readable cues, not one cue per word. This skill groups the words into cues and
emits valid SRT.
Inputs
A job folder path whose input.json already contains measured word timings
(run elevenlabs-tts first). Each word is
{word, start, end} under subtitles[].words[].
Usage
bash skills/subtitle-srt-export/scripts/export_srt.sh
Example (the bundled demo job):
bash skills/subtitle-srt-export/scripts/export_srt.sh examples/demo-job
Cue grouping rules
A new cue starts when any of these is true:
- the previous word ended a sentence (
.,?,!) - the current cue would exceed ~3.5 seconds
- the current cue would exceed ~10 words
Tune the $maxdur / $maxwords constants at the top of the jq block to taste.
Output
Writes /subtitles.srt and prints:
Wrote /subtitles.srt with cues from words
On a missing-timings precondition failure it prints an ERROR: line and exits
non-zero — run elevenlabs-tts first.
相关技能
Re-anchor a long-form video timeline to the MEASURED voiceover. After TTS runs, the timeline segment start/end times are still the PREDICTED (155 wpm) values...
Generate or align SRT subtitles from audio using this repository. Use when the user asks for subtitle generation, transcript-to-audio alignment, timing clean...
16 次安装1 星标
Generate the storyboard images for a short-form video job. Walks the image_prompts[] array from a job's input.json, calls Google's Gemini image model to rend...
Turn final short-video scripts into ready-to-edit voiceover audio for TikTok, Reels, YouTube Shorts, product reviews, hook lines, explainers, and ads. This AI voice over generator, AI voice reader, and text-to-speech voiceover workflow makes scripts speakable, helps choose a suitable voice and supported language, and tunes pacing, pauses, names, numbers, brands, and pronunciation. Review the current price estimate, create MP3 voiceover audio, compare the returned duration with the target edit, and place the narration into short-form video, captions, avatar, lip-sync, and publishing workflows.
ElevenLabs scribe_v1 speech-to-text with auto language detection and optional speaker diarization. Suitable for subtitles, transcription, and meeting notes. ElevenLabs scribe_v1 语音转文字，支持自动语种识别与说话人分离，适合字幕、转录与会议记录。
8 次安装