- Home
- Skills
- Writing & Content
- ElevenLabs SSML Corpus Review
Works with the AI Agents you already use
ElevenLabs SSML Corpus Review
Inventory unsupported SSML elements across a script CSV corpus, preserve original text, and produce per-script review tasks for dictionary, audio-tag or editor handling without pretending timing equivalence.
$5
ElevenLabs SSML Corpus Review
Example session with this skill installed
Illustrative invocation for the seller's synthetic acceptance fixture: review script CSV containing break and phoneme SSML fragments for the declared Eleven model. Keep original text and return review tasks; do not synthesize or convert speech.
- Read your context and instructions
- Compiled the elevenlabs ssml corpus
Seller demonstration — synthetic fixture
This is the complete recorded JSON from the seller's local CLI replay of the exact accepted ZIP, version 0.2.0. It is not a live account export or a verified Agensi-hosted agent session. The invocation above is illustrative; the output below is the preserved actual seller test result.
Complete observed output
{
"coverage": {
"scripts": {
"blank_record_end_lines": [],
"bytes": 115,
"preamble_blank_record_end_lines": [],
"preamble_record_end_lines": [],
"reader": "table",
"record_end_lines": [
2,
3
],
"records": 2,
"source_complete": "NOT_VERIFIED"
}
},
"details": {
"markup_scripts": 2,
"scripts": [
{
"audio_equivalence": "NOT_TESTED",
"automatic_conversion": false,
"directives": [
{
"attributes": {
"time": "1.5s"
},
"direct_text": "",
"review_action": "Review intended pause; use supported audio tags or edit rendered audio; duration equivalence unverified.",
"tag": "break"
}
],
"original_text": "Hello <break time=\"1.5s\"/> world.",
"script_id": "001"
},
{
"audio_equivalence": "NOT_TESTED",
"automatic_conversion": false,
"directives": [
{
"attributes": {
"alphabet": "ipa",
"ph": "kæt"
},
"direct_text": "cat",
"review_action": "Review dictionary phoneme/alias rule; alphabet/model support and pronunciation require listening.",
"tag": "phoneme"
}
],
"original_text": "<phoneme alphabet=\"ipa\" ph=\"kæt\">cat</phoneme>",
"script_id": "002"
}
],
"target_model": "eleven_v4"
},
"findings": [
{
"code": "SSML_DIRECTIVES_REQUIRE_REVIEW",
"count": 1,
"level": "REVIEW",
"script_id": "001"
},
{
"code": "SSML_DIRECTIVES_REQUIRE_REVIEW",
"count": 1,
"level": "REVIEW",
"script_id": "002"
}
],
"input_sha256": {
"policy": "60419c5380c19a10ca07b72cea4353e0b77a7132efa60bed0d47c54e572a6171",
"scripts": "ca061498a3214db0df60d8b058901bc44ba6a2e19cd52f08037488e2948cd065"
},
"limitations": [
"No audio generated or listened to; plain text CLEAR is not audio/voice consistency approval.",
"No conversion, phoneme translation, duration preservation or unsupported model guarantees."
],
"schema_version": "local-audit-result@1",
"status": "REVIEW"
}
Next step
Review the scoped findings and coverage before any manual platform action. CLEAR only covers the documented local rules; input completeness, platform acceptance and business correctness are not certified.
Connects securely to your tools. The creator never sees your data.
About this skill
Inventory unsupported SSML elements across a script CSV corpus, preserve original text, and produce per-script review tasks for dictionary, audio-tag or editor handling without pretending timing equivalence.
For: Commercial voice-over producers moving saved scripts to Eleven v3/v4.
Input: UTF-8 CSV with unique script_id and text; explicit target eleven_v3/eleven_v4. Well-formed XML fragments only; DTD/entities/unknown malformed markup rejected.
Included: original Python source, companion SKILL.md, runnable synthetic examples, policy guide, README and a file-hash manifest. Runs locally on Windows 10/11 with Python 3.11+; no third-party runtime dependencies. Use the documented source command after extraction. A Python-enabled agent is needed for Skill execution; installing the Skill alone does not include the complete source folder. Your AI assistant may have separate costs.
Output: a JSON audit with CLEAR, REVIEW, BLOCK or INVALID_INPUT, findings, scoped details, coverage and input hashes. The audit never edits inputs, uploads data, calls a paid API or executes a platform change. CLEAR covers only the documented rules and is not platform approval or verification that your export is complete.
Limits: Does not convert phonemes, regenerate speech, fix accents, measure audio timing, or prove tag performance. Paid demand is particularly uncertain. Input files are bounded to 1 MiB each; other structural limits are in the included README. Windows local regular files only; links, network drives, cloud placeholders and archives as audit inputs are unsupported.
Compare before buying: Free official ElevenLabs Skills and docs, native dictionary editor, search/replace, XML lint, commercial audio editors. These can solve one script or synthesize audio; our narrow hypothesis is corpus-level traceability before costly regeneration.
This package adds: Loss-aware batch inventory with original scripts, directive counts and review actions; not another TTS wrapper or a claimed automatic model upgrade.
One-time download, $5. This is a bounded commercial experiment; sales and time savings are not established.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 7 days ago
- Passed all security checks, Safe to install
Requires