Learning path · Level 4 · Tutorial 3

Generating and Revising Short Narration

Create and revise a short narrated audio asset using an accessible text-to-speech tool and an authorized voice. If audio generation is unavailable, prepare a narration script with pronunciation, delivery, and listening-check instructions without claiming to have verified unheard audio.

  • Level: Intermediate
  • Duration: 18 min
  • 8 steps
Start tutorial Back to the path
Generating and Revising Short Narration

What you will go through

  1. Write a Script for Listening Turn approved library facts into a concise spoken announcement and estimate whether its length is a reasonable starting point for a 30-second delivery. 2 min
  2. Specify Pronunciation Before Generating Identify pronunciation-sensitive words in the Cedar Library narration and record clear plain-English notes before generation. 2 min
  3. Plan Pace, Pauses, and Emphasis Plan pace, pauses, and emphasis for the Cedar Library narration while keeping delivery specifications separate from the controls a tool actually supports. 2 min
  4. Choose a Voice You Are Authorized to Use Select a permitted synthetic voice for the Cedar Library announcement and distinguish authorization from mere access to a recording. 2 min
  5. Generate a Draft or Complete the Specification Create Draft 1 in an available text-to-speech tool, or preserve a complete text-only narration specification without claiming unperformed audio checks. 3 min
  6. Listen for Specific, Recordable Problems Evaluate available narration with focused listening passes and an evidence-based log, while keeping unavailable audio checks explicitly unverified. 3 min
  7. Revise One Phrase and Check Its Neighbors Make a focused narration correction while preserving unaffected content and recording continuity checks honestly. 3 min
  8. Complete and Review Your Narration Package Combine a revised Cedar Library narration or a generation-ready specification into a coherent package, with route-specific readiness supported by evidence. 3 min

What you will learn

  • Prepare a short spoken script with pronunciation notes, pauses, and a target duration.
  • Choose an authorized synthetic voice and identify when explicit permission would be needed for a person's voice or recording.
  • Generate narration using an available free-tier feature, or complete a text-only narration specification.
  • Check available audio for pronunciation, intelligibility, pacing, and unwanted sounds, then revise a faulty phrase; if audio is unavailable, document the checks that remain unperformed.

Before you start

  • Building and Refining a Task Prompt
  • Writing and Summarizing Without Changing Meaning
  • Applying Basic AI Safety Checks

Step 1 of 8

Write a Script for Listening

Turn approved library facts into a concise spoken announcement and estimate whether its length is a reasonable starting point for a 30-second delivery.

Make the message easy to hear

Write for listening

A listener cannot scan backward as easily as a reader. Shape a short announcement for the ear: use familiar words, short clauses, and clear sentence boundaries.

For Cedar Library, the practical purpose is simple: help first-time visitors locate and use the new self-service return kiosk. Keep every approved fact, but present one action at a time.

Structure makes spoken information easier to follow

A visual comparison between dense prose and short phrase groups for a library kiosk announcement.

The phrase-grouped version gives listeners clear places to process one idea before the next.

Use word count as a starting estimate

Estimate before playback

Use the planning range, not a promise:

  • Target duration: 30 s
  • Acceptable range: 25–35 s
  • Initial planning rate: 130–150 wpm
  • Useful starting length: roughly 55–65 spoken words

Estimate from word count and rate, then allow extra time for pauses. At 130–150 wpm, 55–65 words take about 22–30 s before pauses. This makes the range plausible for a 30 s announcement, but it is only an Estimated duration. Playback later establishes the Measured duration.

Example

A concise spoken draft

Draft 1

Spoken script
Welcome to Cedar Library. If you are visiting for the first time, you can use the new self-service return kiosk near the main entrance. Please have your library ID ready before you begin. Follow the on-screen instructions as you use the kiosk. If you need help at any point, staff at the service desk can help.

This draft has 56 spoken words. Its short sentences make each action easier to follow. It is a starting point for the 30 s target, not proof of the final duration.

Tip

Keep script and annotations separate

At this stage, write only the words the listener should hear. Pronunciation notes and Delivery notes are added separately in later steps; they are not spoken words.

Practice: adapt without adding facts

Turn dense prose into a listening script

Rewrite the dense excerpt below as a spoken announcement of 55–65 words for first-time Cedar Library visitors. Preserve its facts and add none.

“Cedar Library wishes to inform visitors that its new self-service return kiosk, which is situated near the main entrance, may be used by visitors who should ensure that they have their library ID available and then follow the instructions displayed on the kiosk screen, while staff located at the service desk are available to provide help.”

After your rewritten Spoken script, state its word count and explain in one sentence why that length is a plausible starting point for the 30 s target. Do not claim a measured duration.

Write at least 250 characters (0/250).

Step 2 of 8

Specify Pronunciation Before Generating

Identify pronunciation-sensitive words in the Cedar Library narration and record clear plain-English notes before generation.

Identify words that need a note

Find pronunciation-sensitive words

Before generating narration, scan for words that a voice may handle unexpectedly. Three common flags are:

  • a name, such as Cedar
  • an abbreviation, such as ID
  • a word whose sound changes with its meaning, such as read

A pronunciation note records the intended sound. It does not replace the word in the Spoken script, change the spelling of a factual name, or prove that later audio will say it correctly. Playback is needed to verify generated sound.

Keep the written word; add a sound note

A three-lane diagram connects Cedar, ID, and read used as an instruction to their intended plain-English spoken forms.

The script spelling stays intact. A separate note states the intended sound.

Working pronunciation examples

Use plain-English respelling, not technical phonetic codes. The final row is a separate working example: the word read does not occur in the established Draft 1 script.

Write an intended sound with an arrow while preserving the original written form.
Sensitive itemWhy check it?Pronunciation note
CedarNameCedar → SEE-der
IDAbbreviationID → eye dee
read (instruction)Meaning affects soundread (instruction) → REED

Hear the intended sounds

Listen to this tutorial example of the intended English (US) sounds. It demonstrates an intended pronunciation only; it is not evidence about your own narration.

Welcome to Cedar Library. Please have your library ID ready. In this separate example, please read the on-screen instructions.

Attach notes without changing the script

Use a reliable source for real names

For a real person’s or organization’s name, do not turn a confident guess into a pronunciation note. Use a reliable reference, preferably one provided by the person or organization concerned. If you cannot confirm it, record that the pronunciation needs confirmation.

Whether a note comes from a reliable source or a working example, it states an intention for generation. It is not evidence that a tool has produced the intended sound.

Example

Keep Draft 1 unchanged

Draft 1

Spoken script
Welcome to Cedar Library. If you are visiting for the first time, you can use the new self-service return kiosk near the main entrance. Please have your library ID ready before you begin. Follow the on-screen instructions as you use the kiosk. If you need help at any point, staff at the service desk can help.

Pronunciation notes
Cedar → SEE-der
ID → eye dee

Delivery notes
Not specified yet. Delivery planning is added separately in the next step.

Separate pronunciation example
If a script uses read as an instruction, write: read (instruction) → REED.

Tip

Keep script and notes separate

Only words the listener should hear belong under Spoken script. Keep sound guidance under Pronunciation notes. This makes the intended sound reviewable while preserving the approved Cedar Library facts and the established Draft 1 wording.

Apply the notation

Check your pronunciation specification

Match each written item to its intended plain-English spoken form.

Match written items to intended sounds

Match every item to its intended spoken form.

Tap an item and then its matching pair.

Step 3 of 8

Plan Pace, Pauses, and Emphasis

Plan pace, pauses, and emphasis for the Cedar Library narration while keeping delivery specifications separate from the controls a tool actually supports.

Use phrase boundaries deliberately

Plan for listening, not visual scanning

Listeners receive an announcement in time: they cannot scan backward as easily as readers can. A moderate pace, clear phrase endings, and a few purposeful pauses make the directions easier to follow.

Use natural boundaries: after the welcome, after the kiosk location, and before the help offer. Emphasize one useful item, such as “library ID,” rather than stressing every important-looking word. Too much emphasis can make a calm and welcoming announcement distracting.

A phrase timeline is a planning aid. It shows where speech groups and pauses belong; it is not a measured audio waveform or proof of duration.

Phrase timeline

A timeline of five spoken phrase blocks, with short gaps between blocks for pauses and one highlighted phrase for restrained emphasis.

Treat each block as one spoken phrase. The gaps suggest planned pauses; the highlight suggests a single emphasis point.

Separate the plan from tool controls

Write delivery instructions separately

Keep the narration materials in separate sections. The speech-input field normally receives only the words under Spoken script. Pronunciation notes and Delivery notes remain outside that field unless the chosen tool explicitly documents an instruction field or notation it supports.

Delivery notes
After "Cedar Library": [pause: 0.5 s]
After "main entrance": [pause: 0.5 s]
[emphasis: library ID]
Tone: calm and welcoming.

Start planning at 130–150 wpm. Aim for a target duration of 30 s, with an acceptable range of 25–35 s. Pauses add time. This is an Estimated duration, not a Measured duration: playback is required later to establish the actual length.

Match the plan to the tool

A bracketed delivery note is a specification, not a universal command. Check what the selected tool actually offers before using a setting or putting any notation into an instruction field.

A desired result and an available control are not the same thing.
Section or needWhat belongs thereSupport to check
Spoken scriptWords the listener should hearSpeech-input field
Pronunciation notesExisting plain-English guidancePronunciation control, if offered
Delivery notesPauses, emphasis, calm and welcoming toneDocumented instruction or notation support
Pace adjustmentA slower or faster delivery planSpeed or rate control
No matching controlClear wording and sentence punctuationDo not invent unsupported commands

Apply a restrained delivery plan

Choose a supported adjustment

A planning estimate for the Cedar Library announcement exceeds 35 s. Its first sentence is also dense. The tool offers a speed slider, but it has no documented instruction field and no documented pause notation. Which plan best supports clarity while using only controls that are actually available?

Step 4 of 8

Choose a Voice You Are Authorized to Use

Select a permitted synthetic voice for the Cedar Library announcement and distinguish authorization from mere access to a recording.

Start with a permitted voice

Choose qualities, not an identity

For the Cedar Library announcement, start with a built-in synthetic voice only after checking that the provider permits this use. A suitable description is: clear, warm, calm and welcoming, using English (US).

Describe qualities, not a person to imitate. For example, request a calm delivery—not “sound like” an identifiable presenter, actor, colleague, or public figure.

The voice choice is separate from the narration text: keep the approved library facts and the intended delivery unchanged while you decide whether the voice is permitted.

Authorization comes before generation

A diagram contrasts a provider-permitted built-in voice and a consented personal voice with an unverified recording that must not be used yet.

A voice can be used only when its permission path fits the intended use.

Check the permission path

Three authorization paths

Use this decision guide before selecting a voice for the Cedar Library announcement.

Check both the source of the voice and the planned use.
Voice sourceWhat to confirmDecision
Built-in synthetic voiceProvider permits the intended useSuitable starting point
Personal voiceExplicit permission covers this useUse only within that permission
Recording you can accessPermission to upload and process it, if applicableDo not assume access is permission

Warning

Access is not authorization

A public recording, a shared audio file, or familiarity with someone’s voice does not establish permission to imitate that person or to upload their recording to a service.

If a recording would be uploaded or processed, authorization for that handling may also be needed. Check the provider’s current terms and the scope of the person’s explicit permission. This is a practical authorization check, not legal advice.

Apply the authorization decision

Choose the authorized next action

Match each situation to the appropriate next action.

Authorization check

Match each voice-selection situation with its appropriate next action.

Match each situation to the appropriate next action.

Tap an item and then its matching pair.

Step 5 of 8

Generate a Draft or Complete the Specification

Create Draft 1 in an available text-to-speech tool, or preserve a complete text-only narration specification without claiming unperformed audio checks.

Assemble the brief

Build one complete brief

A generation brief keeps the narration inputs together before you open a tool. It has five parts: the Spoken script, Pronunciation notes, Delivery notes, intended voice, and duration information.

The brief is not the same as the speech-input field. It preserves the choices for Draft 1 and a later revision. Keep the target duration at 30 s, with an acceptable range of 25–35 s. An Estimated duration supports planning; it is not a Measured duration.

Keep the brief separate from the input field

A diagram separates spoken words sent to a speech-input field from pronunciation, delivery, voice, and duration notes kept in a generation brief.

Send only the Spoken script to the speech-input field unless the tool documents a separate supported instruction field.

Example

Example: plain-text generation brief

Spoken script
Welcome to Cedar Library. If you are visiting for the first time, you can use the new self-service return kiosk near the main entrance. Please have your library ID ready before you begin. Follow the on-screen instructions as you use the kiosk. If you need help at any point, staff at the service desk can help.

Pronunciation notes
Cedar → SEE-der
ID → eye dee

Delivery notes
After "Cedar Library": [pause: 0.5 s]
After "main entrance": [pause: 0.5 s]
[emphasis: library ID]
Intended delivery: calm and welcoming

Intended voice: provider-permitted built-in synthetic voice, English (US)
Target duration: 30 s
Acceptable range: 25–35 s
Initial planning range: 130–150 wpm
Estimated duration: about 25.0 s at 140 wpm, including the planned pauses
Version: Draft 1

This preserves the established 56-word Draft 1. The separate working note read (instruction) → REED is useful only if that word appears in a future script; it is not needed for this Spoken script.

Use the tool without losing the specification

Choose a route from current availability

Use your own accessible text-to-speech tool. Before relying on it, check its current options: an appropriate free-tier text-to-speech feature, a permitted English (US) voice that fits the brief, enough text allowance for the 56-word script, and playback.

If these are available, create Draft 1. If a needed generation feature is unavailable, complete the same brief as a Text-only route instead. This activity does not require an uploaded recording or tutorial media.

What belongs where

Keep the generation brief open while working. Use only controls the tool currently documents; delivery annotations are not universal commands.

Record actual choices so a later revision can retain them.
Brief componentSpeech-input field?What to record
Spoken scriptYesUse the established words exactly
Pronunciation notesOnly with a documented controlKeep in the brief otherwise
Delivery notesOnly with a documented instruction fieldKeep in the brief otherwise
Voice and settingsNoRecord the selected label and available settings
Duration informationNoKeep target and Estimated duration in the brief

Tip

When audio cannot be generated

For the Text-only route, label the route Text-only route and write Audio not generated. Keep the complete brief ready for later generation.

Create a Listening log with the columns Check, Status, Evidence, and Next action. Audio-dependent entries are Not checked until playable audio exists. If you use a hypothetical result while planning, label it Example observation; do not present it as something you heard.

Sequence the available-audio route

Order the Draft 1 workflow

For the Audio route, place these actions in the recommended order.

  1. Select a permitted built-in English (US) voice and record its actual label and available settings.
  2. Check the current free-tier text-to-speech feature, voice option, text allowance, and playback availability.
  3. Generate Draft 1 and confirm that playable audio exists.
  4. Put only the established Spoken script into the speech-input field; keep unsupported notes in the brief.

Record the result honestly

Document your route

Record what you actually did. Do not assess pronunciation, pacing, phrase boundaries, or sound quality yet.

Audio route: state Audio route; report the tool checks you completed, the actual selected voice label and available settings, and whether Draft 1 has playable audio.

Text-only route: state Text-only route and Audio not generated; confirm that the specification contains Spoken script, Pronunciation notes, Delivery notes, intended voice, and duration information. State that audio-dependent Listening log checks are Not checked.

Write at least 80 characters (0/80).

Step 6 of 8

Listen for Specific, Recordable Problems

Evaluate available narration with focused listening passes and an evidence-based log, while keeping unavailable audio checks explicitly unverified.

Make the first and focused passes

Listen in two passes

Use two passes when Draft 1 is available.

First pass: listen for the message. Can a first-time visitor understand what to do, and are all required facts present?

Focused pass: compare what you hear with the Spoken script. Check pronunciation-sensitive words, the library ID, pace, planned pauses, and phrase endings. Notice whether words are omitted, repeated, changed, or cut off.

Listening establishes what was actually produced. The script, Pronunciation notes, and Delivery notes establish only what you intended.

From intended script to observed audio

A two-part diagram shows a broad first listen followed by a focused comparison of audio, script, and duration.

The first pass checks the whole message. The focused pass checks specific, recordable details.

Tip

Evidence rule

Do not treat a transcript, pronunciation note, or delivery plan as proof that the audio is clear. These materials guide the check, but only playback can reveal pronunciation, intelligibility, pace, silence, phrase boundaries, or unwanted sound.

Use an honest Listening log

Log only what your route supports

Record the actual result, not the earlier estimate. For an Audio route, read or measure the playback and label it Measured duration; compare it with the 25–35 s acceptable range around the 30 s target. An Estimated duration is not a listening result.

Also listen for clicks, hiss, unexpected sounds, and clipped starts or endings. You do not need specialist audio software: a phrase reference, approximate playback time, or Measured duration is useful evidence.

For a Text-only route, write Audio not generated. Every audio-dependent check remains Not checked until playable audio exists.

Audio route: example Listening log

This is a complete example for an Audio route. Each observation is hypothetical and is labeled Example observation; it is not a finding about your Draft 1.

An Audio route can contain heard evidence and a Measured duration.
CheckStatusEvidenceNext action
Message and wordsPassExample observation: all required facts heard onceNo change
PronunciationNeeds revisionExample observation: around 9 s, ID is unclearRecord the faulty phrase
Measured durationPassExample observation: 29.4 sNo change
Unwanted soundsPassExample observation: no click, hiss, or clipped ending heardNo change

Text-only route: example Listening log

This is a separate complete example for a Text-only route. It does not claim that any audio was heard or measured.

Without playable audio, audio-dependent checks remain unperformed.
CheckStatusEvidenceNext action
Message and wordsNot checkedAudio not generatedListen after generation
PronunciationNot checkedAudio not generatedListen after generation
Measured durationNot checkedAudio not generatedRead or measure playback after generation
Unwanted soundsNot checkedAudio not generatedListen after generation

Practice: diagnose a specific problem

Identify the audible issue

Listen and choose an option.

Listen to this controlled tutorial excerpt. Which audible issue should be recorded? This is an excerpt, not a full narration: do not assess its duration or completeness.

Transcript

Welcome to Cedar Library. The return kiosk is near the main entrance. Please have your library ID ready. Please have your library ID ready. Follow the on-screen instructions.

Apply the checklist to your route

Record evidence without overclaiming

Create Listening log entry or entries in this format:

Check | Status | Evidence | Next action

Audio route: state Audio route. Record at least one check you actually heard or measured in your own Draft 1. Include an approximate playback time or Measured duration when relevant. Do not use a tutorial sample as your evidence.

Text-only route: state Text-only route and Audio not generated. Record an audio-dependent check as Not checked. Then add a separate, clearly labeled diagnosis of this case: Example observation: at approximately 7 s, the library ID sentence is heard twice. This diagnosis is not evidence from your narration.

Write at least 40 characters (0/40).

Step 7 of 8

Revise One Phrase and Check Its Neighbors

Make a focused narration correction while preserving unaffected content and recording continuity checks honestly.

Limit the revision and protect the rest

Make one purposeful change

A controlled revision changes one identified problem—not the whole narration by accident. Start with the faulty phrase and the correction it needs. Then explicitly protect what must remain the same: the approved Cedar Library facts, the selected authorized voice, the recorded Draft 1 settings, and delivery choices that are not part of the problem.

Match the correction to the fault:

  • A pronunciation note was ineffective: use a documented pronunciation control, if the tool provides one; retain the existing note as the specification record.
  • A phrase is rushed: revise only its wording or pause plan.
  • A click, hiss, clipping, or other unwanted sound occurs: regenerate the affected material.

If the tool supports phrase replacement, include enough surrounding words to preserve the transition. If it does not, regenerate the complete short narration with only the intended input change. You do not need an audio editor or manual splicing.

Check the joins, not only the corrected words

A three-part narration timeline shows a revised middle phrase connected to the phrase before and after it, with both boundaries marked for inspection.

The revision is only the center segment, but continuity must be checked at both neighboring boundaries.

Use a revision record

Record the reason and the scope

Keep the revision record short and traceable. It states what changed and what was intentionally preserved. Do not describe a listening result unless you actually heard your own Draft 1 or Revision 1.

For an Audio route, listen across the phrase before the change, the revised phrase, and the phrase after it. Check voice character, loudness, pace, silence, and clipped word boundaries. Then listen to the full Revision 1 and read or measure its Measured duration; a small revision can cause duration drift outside the 25–35 s range.

For a Text-only route, state Audio not generated. Record the intended correction, but mark all audio-dependent checks as Not checked. Written instructions can specify an expected result; they cannot verify it.

Example

Example: revise an ineffective pronunciation note

Revision 1

Faulty phrase: “library ID” — Example observation: despite the existing note ID → eye dee, ID was heard as one word rather than as letter names.

Change: Keep the existing Pronunciation notes unchanged. In a tool that offers a documented pronunciation control, configure ID to be spoken as separate letter names, then regenerate the affected phrase with surrounding context or regenerate the complete short narration.

Keep unchanged: The Spoken script wording and all approved Cedar Library facts; the calm and welcoming Delivery notes; the selected voice and the recorded Draft 1 settings.

This example does not claim that your narration has this fault. If your tool has no documented pronunciation control, record the correction as pending rather than inventing a command or claiming it was fixed.

Practice selecting the correction

Choose a fault-specific correction

Match each Draft 1 problem to its most suitable first correction.

Tap an item and then its matching pair.

Document your revision and pending checks

Apply one controlled revision

Create a Revision 1 record for one issue from your own Draft 1. If you have no observed fault or no audio, use this labeled case instead: Example observation: despite the existing note ID → eye dee, ID was heard as one word rather than as letter names. Do not present that example observation as something you personally heard.

Include these fields:

  • Revision 1
  • Faulty phrase
  • Change
  • Keep unchanged
  • Audio route or Text-only route
  • Listening log with: Check | Status | Evidence | Next action

Your log must cover the boundary before the revised phrase, the boundary after it, and the full revised narration including Measured duration. Use Audio route only if you generated Revision 1 and listened to it. Otherwise use Text-only route, write Audio not generated, and mark audio-dependent checks Not checked. Preserve the approved facts, authorized voice, and unaffected Draft 1 settings.

Write at least 250 characters (0/250).

Step 8 of 8

Complete and Review Your Narration Package

Combine a revised Cedar Library narration or a generation-ready specification into a coherent package, with route-specific readiness supported by evidence.

Assemble the final record

One package, clear evidence

A final narration package keeps the request, result, and evidence together. It is more than an audio file or a script.

Keep these parts identifiable:

  • Spoken script — final words only; no pause labels or sound guidance.
  • Pronunciation notes — intended sounds for sensitive words.
  • Delivery notes — pace, pauses, emphasis, and intended delivery.
  • Voice choice and authorization basis — why the selected voice may be used.
  • Draft 1 and Revision 1 — one controlled, documented change.
  • Listening log — what you actually checked, with evidence and a next action.

Preserve the approved Cedar Library facts. A revision may improve clarity, but it must not add services, hours, fees, or unsupported details.

Package structure

Use the same package structure whether you generated audio or prepared for later generation.

A diagram links script, pronunciation, delivery, permitted voice, revision, and listening evidence to either a checked audio outcome or a text-only outcome with pending checks.

The route changes the evidence you can claim; it does not remove the need for a complete specification.

Keep specification sections separate

Use these exact headings in your record:

Spoken script
Only the final announcement words belong here.

Pronunciation notes
Include applicable guidance, such as Cedar → SEE-der and ID → eye dee.

Delivery notes
Include the calm and welcoming delivery, planned pauses, and emphasis.

Retain the target duration of 30 s, acceptable range of 25–35 s, and initial planning range of 130–150 wpm. Write an Estimated duration before playback. Write a Measured duration only after actual playback is available.

Match the status to the evidence

Choose the honest route

Readiness is an evidence-based claim. An audio draft is not ready merely because a tool produced sound. A text-only specification is valuable, but it is ready for generation—not proof that the narration sounds correct.

What each route can support

Use the route that matches what you actually have available.

A final status must be supported by the evidence in the Listening log.
CategoryAudio routeText-only route
ProducedRevised audio is availableAudio not generated
Pronunciation and intelligibilityRecord actual listening evidenceMark Not checked
Pace and phrase boundariesRecord actual listening evidenceMark Not checked
Unwanted soundsRecord actual listening evidenceMark Not checked
DurationRecord Measured durationKeep Estimated duration only
Final statusReady only when checks have Pass evidence; otherwise reviseReady for generation, not a verified audio asset

Warning

Avoid unsupported verification

Do not turn an intention into an observation. “The pronunciation note requests eye dee” is a specification. “ID was heard as separate letter names” requires listening to your own audio. If you did not generate audio, state Audio not generated and leave audio-dependent checks as Not checked.

Present an evidence-based final package

Final application: present your package

Create your final narration package below. Include the exact headings Spoken script, Pronunciation notes, and Delivery notes. Paste or summarize your final script and notes; do not add facts beyond the approved Cedar Library announcement.

State the route, target duration, and either an Estimated duration or a Measured duration. State the authorization basis for the selected voice. Include Draft 1 and Revision 1. For Revision 1, use Faulty phrase, Change, and Keep unchanged.

Complete a Listening log using Check | Status | Evidence | Next action. Use only Pass, Needs revision, or Not checked as Status values. State a final status that matches your evidence. Use Audio route only when you have revised audio and actual listening evidence. Otherwise use Text-only route, write Audio not generated, and never describe unheard audio as verified.

Write at least 400 characters (0/400).

Finish with an honest readiness claim

Summary

Final review

A responsible final review distinguishes specification from verification.

  • The package combines the final script, pronunciation notes, delivery notes, authorized voice choice, and revision record.
  • Audio route: use actual listening evidence and Measured duration to support readiness.
  • Text-only route: write Audio not generated and keep audio-dependent checks as Not checked.
  • A controlled revision identifies the faulty phrase, the change, and what remains unchanged.

Tutorial completed

Congratulations! You completed: Generating and Revising Short Narration

You completed the narration package. You can now present either a revised, evidence-checked narration or a generation-ready text-only specification without overstating what has been verified.

Download the App now to have access to + 5000 free courses, exercises, certificates and lots of content without paying anything!

  • 100% free online courses from start to finish

    Thousands of online courses in video, ebooks and audiobooks.

  • More than 60 thousand free exercises

    To test your knowledge during online courses

  • Valid free Digital Certificate with QR Code

    Generated directly from your cell phone's photo gallery and sent to your email

Cursa app on the ebook screen, the video course screen and the course exercises screen, plus the course completion certificate