elevenlabs-stt
Skill6 starsChanged 23 days agoMITView on GitHub
Use when transcribing speech with ElevenLabs Scribe in any language (Hebrew, English, mixed) - a meeting, talk, call, podcast, voice note or video, a YouTube link, a transcript.md with timestamps and speakers, or SRT subtitles from audio. Also for speaker labels (diarization), word timestamps, keyterms for names and jargon, numbers as digits or words, filler removal, PII redaction, async webhooks for long files, and live or streaming transcription (scribe_v3_live, the realtime WebSocket, turn detection, voice agents). Use when moving from scribe_v2 or scribe_v2_realtime to Scribe v3, when English terms come out in Hebrew letters, when a request fails with "numbers_format is not supported with scribe_v2", "Requesting additional formats must have diarization", or a 500 service_unavailable, or whenever the user says Scribe, 11labs or ElevenLabs speech-to-text, STT or transcription, even without naming ElevenLabs.
- Reads credentials
2 other projects use a version of this.
Adopted by
Most recently taken up by
- AngelescuCiprian/claude-skillstheir own version, first seen Aug 22, 2026
- nikunj812/claude-skilltheir own version, first seen Jul 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- AngelescuCiprian/claude-skills★ 0same name
- nikunj812/claude-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository29 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skills/prompt-engineer/references/claude.mdCLAUDE.md
- brand-assetsSkill
- brand-systemSkill
- codex-reviewSkill
- deck-to-videoSkill
- elevenlabs-ttsSkill
- excalidraw-diagramsSkill
- gdoc-syncSkill
- generating-video-clipsSkill
- gsheetsSkill
- gslides-syncSkill
- image-generationSkill
- namecheap-domainsSkill
- narrating-pptxSkill
- office-renderSkill
- pair-intelligenceSkill
- presentation-generatorSkill
- producing-animated-episodesSkill
- prompt-engineerSkill
- proton-passSkill
- reviving-sessionsSkill
- self-presenting-decksSkill
- tavily-extractSkill
- tech-design-docSkill
- viewing-videosSkill
- writing-plainlySkill
- writing-project-instructionsSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 134
Update anything keyed by model id: billing or metering price tables (a missing row can bill the new model at zero), dashboards, tests, and …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ElevenLabs STT (Scribe v3)
Overview
Audio → text with ElevenLabs Scribe. Scribe v3 (scribe_v3, released 2026-11-10) is the default for recorded audio and Scribe v3 Live (scribe_v3_live) for streams. ElevenLabs reports 27% fewer word errors than v2 across 102 languages, diarization up to 32 speakers, and the same price as v2. Upgrading is a model_id change; every v2 parameter still works.
You cannot hear the audio. So check what a machine can check: the language detected, the speaker count, stretches with no words (dropped speech), and, for Hebrew or Arabic speech, how many English terms came back in Latin script. Then tell the human what to spot-check by ear, and never say a transcript is correct because it reads well.
Requirements: an API key with Speech to Text permission, Python 3 (transcribe.py is stdlib-only), ffmpeg for video input or cut ranges, and uv for the live test bench.
The key. The scripts read ELEVENLABS_API_KEY, else the first line of ~/.config/elevenlabs/api_key (chmod 600). If the user keeps keys in a notes file, load it inside the command substitution, matched by the line's label, never by line number. Never print the key.
Choose the model
| Model | Use for |
scribe_v3 | Default for any recorded audio or video: POST /v1/speech-to-text |
scribe_v3_live | Live audio (voice agents, captions, dictation): the realtime WebSocket, see Live transcription |
scribe_v2 | A/B against v3 on your own audio before moving a pipeline that depends on how its output looks, or to keep one consistent mid-project. Still available |
scribe_v2_medical | Clinical audio, batch only |
scribe_v2_realtime | Old realtime model; the WebSocket's default if you omit model_id |
scribe_v1 | Deprecated |
model_id is required on the batch endpoint. On the WebSocket it defaults to scribe_v2_realtime, so always send it.
Transcribe a file or a link
python3 <this-skill-dir>/scripts/transcribe.py talk.mp4 --out transcripts/ --language he \
--keyterms "Claude Code,EnterpriseRAG-Bench,pass@k,sandbox,agent" --title "Talk title"
python3 <this-skill-dir>/scripts/transcribe.py --url https://youtu.be/VIDEO_ID --out transcripts/ --name talk
python3 <this-skill-dir>/scripts/transcribe.py meeting.m4a --out t/ --speakers 7 # bound the speaker count
python3 <this-skill-dir>/scripts/transcribe.py meeting.m4a --out t/ --start 41:10 --end 43:00 # re-check one passage
python3 <this-skill-dir>/scripts/transcribe.py --from-json t/meeting.scribe.json --out t/ # re-render, no API callIt writes <name>.scribe.json (the raw response: keep it, because re-rendering is free and re-transcribing is not) and <name>.transcript.md, one [mm:ss] speaker_N: text paragraph per speaker turn or pause. Then it prints the length, the language, the speaker count, the Latin-script word count, any 45 s or longer stretch with no words, and the cost at list price. --help lists every option. Video is cut to a mono 16 kHz mp3 first, so a 1 GB mp4 uploads as a few MB.
--urltakes hosted media and YouTube or TikTok links (source_url), so nothing is downloaded. A 21.5-minute YouTube talk came back in 120 s (2026-11-13).- Speed: v3 returned 3 minutes of audio in 26-34 s and 5 minutes in 36 s, about 40% faster than v2 on the same clips.
- Speaker labels are voice clusters, not names. Map
speaker_0to a person from context or a participant list, and say which mappings are guesses. Similar voices merge:--speakers Nsets an upper bound when you know the headcount. - Gap warnings mean either silence or speech the model dropped. If people talk there, transcribe that range with
--start/--endand splice it in, noting that in the transcript's header.
To call the API directly: it is multipart form data. A list parameter (keyterms, secondary_language_codes) is sent as the same field repeated, once per value. additional_formats and entity_detection take a JSON string.
Parameters that change the result
| Parameter | What it does | Notes |
diarize, num_speakers | speaker labels, up to 32 | num_speakers is an upper bound; diarization_threshold tunes it when no count is given |
keyterms | bias toward names, products, jargon | up to 1,000 terms, ≤50 chars and ≤5 words each; +$0.05/h; over 100 terms bills at least 20 s |
language_code | main language (ISO 639-1/3) | auto-detected when omitted; set it for short or one-language clips |
secondary_language_codes | other languages in the audio | v3 only, and send language_code with it: alone it returned HTTP 500 twice |
numbers_format | digits (42) or words (forty-two) | v3 only (v2 returns 400); unset = the model chooses |
no_verbatim | drops fillers, false starts, sounds | good for subtitles and summaries; it also changes how Hebrew transcripts write English terms (below) |
tag_audio_events | (laughter), (applause) in the text | the script sends false unless --events |
transcript_edit | a plain-English rewrite rule, result in edited_transcript | +$0.07/h, experimental; not with entity detection or multichannel. Unreliable for changing scripts or spellings (below) |
entity_detection, entity_redaction | find or mask PII, PHI or PCI (65 entity types) | entity_redaction_mode sets the placeholder |
detect_speaker_roles | labels speakers agent / customer | needs diarize; +10% |
use_multi_channel | one speaker per channel, up to 5 channels | each channel is billed at the full length |
webhook | returns at once; the result goes to the workspace's STT webhook | for long files and pipelines; webhook_metadata echoes your job id |
Full parameter list, response shape, limits and errors: references/api.md.
Hebrew and other mixed-language speech
Measured on 2026-11-13 on a 3-minute public Hebrew tech talk and a 5-minute, 7-person Hebrew engineering meeting. Full numbers are in references/hebrew.md.
- v3 can write English terms in Hebrew letters. On the talk, v2 kept 37 English terms in Latin script (
ה-agent,state machine) and plain v3 kept 2-6: it wroteהאייג'נט,סטייט מאשין,סנדבוקס. On the meeting, v3 kept most terms in Latin but still wroteפרוקסי,פייתון, and Title-Cased the rest (ה-Agent,Web Search).language_code=hewithsecondary_language_codes=endid not change it. - Keyterms fix recognition, and sometimes the script too. Keyterms for the terms actually said fixed misheard names and jargon (
merge queue,SSM). On the talk they also brought Latin script back for 26-33 words. On the meeting they didn't:Pythonandproxy, even when listed, stayedפייתוןandפרוקסי. Terms that aren't said do nothing, so build the list from the agenda, slides, participants and tool names. no_verbatimrestored Latin script on both clips: 34 words on the talk, and on the meetingPython/proxymatched v2. It also drops fillers and false starts, so use it (--clean) when a clean, searchable transcript matters more than a verbatim record. For a verbatim record, use keyterms and accept some Hebrew-script loanwords, or keep v2.- Don't use
transcript_editto change the script. "Write English terms in Latin script" romanized the Hebrew itself (O-ma kora merge-shel). - Read the
Latin-script wordsline the script prints. If English terms are expected but the count is near zero, re-run with keyterms, then with--clean. - The word errors that remain are names and jargon in both versions (v3 wrote "evils" for evals, and v2 "Judge" where v3 heard "Jaja"). Keyterms fix those too. Before switching a pipeline whose readers search transcripts for English terms, A/B v2 and v3 on one real recording.
- Scribe does not translate: speech comes back in the language it was spoken in.
Subtitles (SRT)
python3 <this-skill-dir>/scripts/transcribe.py talk.mp4 --out subs/ --srt --keyterms "..." [--srt-chars 42 --srt-secs 6] [--clean]This asks the API for SRT through additional_formats with max_characters_per_line and max_segment_duration_s. Measured result: at most 2 lines, 42 characters per line and 5.94 s per cue. The API refuses additional formats unless diarization and word timestamps are on; the script sends both. The API's SRT also contains sub-second fragment cues, such as a lone word for 0.18 s. The script merges each one into a neighbour when the result still fits the limits. A fragment that can't merge whole is re-split with its neighbour at a word boundary (a sentence end when possible), using the response's word timestamps. The API's original is kept as <name>.api.srt. On the test talk this did 4 merges and 2 re-splits, leaving no lone-word cue. Don't write your own cue splitter. --clean (no_verbatim) reads better on screen, at the cost of a verbatim record. Check reading speed: fast speakers can push a cue past 17-20 characters a second, and the fix is shorter wording by hand, not longer cues. To burn the subtitles into a video, see deck-to-video. Other formats: txt, docx, pdf, html, segmented_json (see references/api.md).
Long files and pipelines
- Limits: 5 GB per upload (
source_urlup to 2 GB), 15 hours of audio, at least 100 ms. - Synchronous requests hold the connection until done: give them a long timeout (the script uses an hour).
webhook=truereturns at once with arequest_id, and the result is POSTed to the STT webhook set in the workspace settings, not in the request. Use it when a job queue or a serverless function must not wait.- Speed: a file longer than 8 minutes is split and transcribed in parallel, so long files come back fast. A 42-minute TV episode took about 27 s (22 episodes, v2 and v3 alike).
- Concurrency per plan: Creator 20, Pro 40, Scale and Business 60. A long file uses
min(4, ceil(seconds / 480))of those slots, so on Creator only about 5 long files run at once. Firing 21 episodes at once got 15 HTTP 429s; 4 at a time (xargs -P 4) worked. A 500service_unavailablesaid "You have not been charged" when seen, and the script retries both with backoff. - Word timestamps matched a local Whisper check to within 0.2-0.5 s in normal dialogue. Three cases break them:
- Music: songs and music under dialogue bunch the timestamps, with many words sharing one start and end, and lyrics come back as speech. Guard any sync or alignment logic against runs of identical times.
- Short lines: shouts, crowd noise and mumbled one-word lines ("Okay", "Hey") can be missing.
- Names come back as spoken ("Zoey heart" for Zoe Hart). Pass the cast list as keyterms, or match against subtitles fuzzily.
- Data residency: US by default. EU, India and Singapore environments are Enterprise-only, use separate accounts and keys, and lack
transcript_edit.enable_logging=false(zero retention) is Enterprise-only.
Live transcription (Scribe v3 Live)
wss://api.elevenlabs.io/v1/speech-to-text/realtime?model_id=scribe_v3_live&audio_format=pcm_16000
&commit_strategy=turn_prediction&diarize=true&include_timestamps=true[&language_code=he][&multilang_mode=true]- Auth: an
xi-api-keyheader from a server, or a 15-minute single-usetokenfor a browser (minted server-side; never ship the key to a client). - Send
{"message_type":"input_audio_chunk","audio_base_64":…,"commit":false,"sample_rate":16000}chunks.previous_textis allowed only on the first chunk. - Commit strategies:
manual(the default: nothing is committed until you sendcommit: true),vad(commits after silence; 1.5 s by default), andturn_prediction(v3 Live only: also commits when the model predicts the speaker has finished, so a caller pausing mid phone number is not cut off). - Receive
partial_transcriptfor display, thencommitted_transcript,committed_transcript_with_timestampsanddiarized_transcript. The diarized message arrives after its committed text. Apply eachspeakers_mergedmessage, in order, to labels you have already shown. - Keyterms live: up to 50 terms of 20 characters each, plus a keyterm library of up to 10,000 (+$0.08/h).
- The official SDKs don't yet support turn prediction, multilingual mode or the keyterm library. Use the raw WebSocket for those.
- Test settings before building:
uv run <this-skill-dir>/scripts/stream_live.py FILE --secs 60 --param language_code=he. It streams a file in real time and prints what the server accepted and returned. Smoke-tested on 2026-11-13 with Hebrew audio: commits, diarization and timestamps all arrived.
Message types, errors and what was measured: references/realtime.md.
Cost (list price, per audio hour)
| Batch (v3, v2, v2 medical) | Live (v3 Live, v2 Realtime) | |
| Base | $0.20 | $0.22 |
| Keyterms | +$0.05 | +$0.08 (and the keyterm library) |
| Transcript edit | +$0.07 | +$0.12 |
Plans include hours (Creator: 110 batch hours a month). Silence is billed. A 1-hour meeting with keyterms costs about $0.25.
Moving a pipeline from v2 to v3
- Change
model_id(scribe_v2→scribe_v3,scribe_v2_realtime→scribe_v3_live). The request parameters carry over. - A/B one representative real recording on both models, comparing names, jargon, and (for mixed-language audio) the script English terms come out in. Add keyterms if v3 transliterates them.
- Update anything keyed by model id: billing or metering price tables (a missing row can bill the new model at zero), dashboards, tests, and the "made with Scribe v2" line in transcript headers.
- Through a third party (Recall.ai's
elevenlabs_async, gateways), check that it passesscribe_v3through before switching. Their docs still showed only v2 on 2026-11-13.
Common mistakes
| Mistake | Consequence | Fix |
Omitting model_id on the WebSocket | silently runs scribe_v2_realtime | send scribe_v3_live |
| Habits carried over from v2 for Hebrew tech talk | English terms come back in Hebrew letters | keyterms, then --clean; check the Latin-script count |
numbers_format or secondary_language_codes on v2 | HTTP 400 | v3 only |
secondary_language_codes without language_code | HTTP 500 | send both |
additional_formats without diarization | HTTP 400 "must have diarization and timestamps enabled" | diarize=true, timestamps_granularity=word (the script forces them) |
| Building SRT cues by hand from word timestamps | cues too long or split mid-word | additional_formats srt with max_characters_per_line and max_segment_duration_s |
transcript_edit used to fix spelling or script | romanized or rewritten text | keyterms; fix names in post-processing |
Treating speaker_0 as a person | wrong attribution | map clusters to names from evidence; say which mappings are guesses |
| Throwing away the raw JSON | paying again to re-render | keep <name>.scribe.json; use --from-json |
| "The transcript is right" | you can't hear it | report gaps, speaker count and Latin-script count; name passages for a human to spot-check |
| Downloading a YouTube video to transcribe it | wasted time and disk | --url (source_url) |
manual commit strategy on a live mic | nothing is committed until you commit | vad or turn_prediction |
Related skills
- elevenlabs-tts: the speech side; its
check_tags_spoken.pyuses Scribe to catch audio tags that were read aloud. - deck-to-video: burns an
.srtinto an mp4. - viewing-videos: what is on screen when the transcript isn't enough (slides, demos, names on badges).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

