Technical Checks
- 44100 Hz sample rate
- mono channel
- 16-bit PCM WAV
- no broken files
Rebuild Checklist
A practical deployment checklist for rebuilding the diphone and TTS system safely
Checklist
Rebuild Checklist
This page provides the practical rebuild checklist for the Bishnupriya Manipuri diphone and TTS workflow, including rule freezing, folder cleanup, audio preparation, validator testing, and deployment readiness.
About this checklist.
Rebuilding a diphone inventory is not just an audio task. It also depends on stable
IPA conversion, phoneme tokenization, diphone sequencing, safe filename mapping,
and validation logic. This page is the operational guide for performing a clean rebuild
without mixing old and new systems.
1. Why a Rebuild Checklist Is Necessary
A rebuild often fails not because of one big mistake, but because several small mismatches accumulate: changed IPA rules, renamed filenames, leftover old files, inconsistent source audio, or incomplete validation.
Main rebuild risks:
- old diphone files remain in the live folder
- safe filename rules changed after some files were already generated
- the validator and TTS engine no longer expect the same filenames
- some recordings were normalized differently from others
- the test list was too small to expose real problems
2. Rebuild Overview
Freeze rules ↓ Back up old system ↓ Create clean output folder ↓ Prepare source audio ↓ Generate / segment diphones ↓ Validate ↓ Fix missing or mismatched items ↓ Deploy
3. Phase 1: Freeze the Rule System
Before rebuilding any audio inventory, make sure the logical system is stable. A rebuild should never begin while pronunciation rules are still shifting.
| Rule Layer | Must Be Frozen? | Why |
|---|---|---|
| IPA conversion rules | Yes | Diphone generation depends on stable IPA output |
| Phoneme tokenization | Yes | Tokenization affects diphone sequence structure |
| Diphone generation logic | Yes | Expected transition units must remain consistent |
| Safe filename mapping | Yes | Validator and playback rely on exact file names |
Do not start a rebuild while still “testing a few possible rules.”
Test and finalize first, then rebuild.
4. Phase 2: Back Up the Old System
Always back up the old diphone folder and related working files before replacing anything.
Recommended backup targets:
- old diphone audio folder
- validator spreadsheet or tracker
- safe filename mapping notes
- recording list and segmentation logs
- current converter files
5. Phase 3: Create a Clean Working Folder
Never rebuild by mixing fresh files into an uncertain old folder. Use a new clean directory for the rebuilt output.
old_live_diphone/ ← keep as backup new_rebuild_diphone/ ← use for current rebuild
This avoids confusion caused by leftover files generated under older rule systems.
6. Phase 4: Prepare Source Audio
The rebuild depends on usable word-level source recordings. Before segmentation, check that source audio is clean and technically consistent.
Speech Checks
- clear pronunciation
- consistent loudness
- trimmed silence
- minimal background noise
Workflow Checks
- word list matches recordings
- recordings have usable filenames
- IPA targets are ready
- segmentation plan is prepared
7. Phase 5: Generate or Segment Diphones
Once source words are prepared, generate expected diphone sequences and segment or extract the corresponding audio units.
Source word ↓ IPA ↓ Phoneme sequence ↓ Diphone sequence ↓ Safe filenames ↓ Segment / save files
Save all new files into the clean rebuild folder, not directly into the live production folder.
8. Phase 6: Run Validator on a Curated Test List
Before full deployment, validate against a small but representative word list.
A good validation list should include:
- simple high-frequency words
- boundary-sensitive words
- cluster-bearing words
- words with rare or difficult transitions
- words from actual dictionary usage
A curated set of 30–50 words is usually enough to expose most system-level mismatches.
9. Phase 7: Review Validation Results
The validator should clearly separate passing items from partial or failing ones.
| Status | Meaning | Action |
|---|---|---|
| Pass | All expected files exist and match | No action needed |
| Partial | Some files exist, some are missing | Rebuild only the missing parts |
| Fail | Major mismatch in rule logic or file set | Stop and review rule consistency |
| Rule Mismatch | Expected and generated outputs disagree | Review converter / filename logic before deployment |
10. Phase 8: Fix Missing or Mismatched Items
Use the validator output as a rebuild target list. Fix only what is wrong, but always check whether the problem is:
- a missing audio file
- a bad segmentation result
- a safe filename mismatch
- a changed IPA rule
- an old leftover file interfering with the test
Do not patch blindly. First identify whether the problem is
audio-related, naming-related, or rule-related.
11. Phase 9: Run the Validator Again
After fixing the detected issues, run the validator again on the same test list. If possible, expand testing to a wider list before deployment.
Initial validation ↓ Fix issues ↓ Repeat validation ↓ Confirm stable results ↓ Deploy
12. Phase 10: Deploy Only After Passing Checks
Move the rebuilt files into the live folder only after the test set behaves consistently.
Deployment-ready conditions:
- shared converter is stable
- safe filename rules are frozen
- validator results are acceptable
- old files are not mixed into the live folder
- word-page TTS matches validator output
13. Full Rebuild Checklist
| Step | Checklist Item | Done? |
|---|---|---|
| 1 | Freeze IPA conversion rules | ☐ |
| 2 | Freeze phoneme tokenization rules | ☐ |
| 3 | Freeze diphone generation rules | ☐ |
| 4 | Freeze safe filename mapping rules | ☐ |
| 5 | Back up old diphone folder | ☐ |
| 6 | Create clean rebuild folder | ☐ |
| 7 | Verify source audio technical format | ☐ |
| 8 | Trim and normalize recordings | ☐ |
| 9 | Generate expected diphone sequences | ☐ |
| 10 | Segment or prepare diphone files | ☐ |
| 11 | Run validator on curated test list | ☐ |
| 12 | Fix missing or mismatched files | ☐ |
| 13 | Run validator again | ☐ |
| 14 | Confirm word-page playback matches validator | ☐ |
| 15 | Deploy rebuilt folder to production | ☐ |
14. Recommended Folder Strategy
/audio/diphone_backup_old/ /audio/diphone_rebuild/ /audio/diphone_live/
Keeping backup, rebuild, and live folders separate makes troubleshooting much easier.
15. Related Archive Pages
Recording Protocol
Review the source recording and normalization workflow.
Validator Workflow
Review validator logic, coverage calculation, and failure types.
Safe Filename Mapping
Review the naming system that must stay stable during rebuild.
Checklist note.
This page is intended to become the operational rebuild guide for the project.
Over time, it can be expanded with printable checklists, downloadable spreadsheets,
and test-word bundles for deployment verification.