| | 1 | |
| | 2 | = Voice building for a new language = |
| | 3 | |
| | 4 | |
| | 5 | == 1. Download xml dump of wikipedia in your language == |
| | 6 | |
| | 7 | == 2. Extract clean text and most frequent words == |
| | 8 | |
| | 9 | == 3. Transcribe most frequent words == |
| | 10 | |
| | 11 | a. Create pronunciation dictionary and train letter-to-sound rules |
| | 12 | b. Minimal NLP components for the new language |
| | 13 | |
| | 14 | == 4. Run feature maker with the minimal nlp components == |
| | 15 | |
| | 16 | == 5. Database selection == |
| | 17 | |
| | 18 | select a phonetically/prosodically balanced recording script |
| | 19 | |
| | 20 | == 6. Manually check/correct transcription of all words in the recording script [Optional] == |
| | 21 | |
| | 22 | == 7. Record script with a native speaker using our recording tool "Redstart" == |
| | 23 | |
| | 24 | == 8. Build an unit selection and/or hmm-based voice with Voice import tool == |
