1. Introduction

Traditional dictionaries provide written definitions and grammatical information. However, modern digital dictionaries can also include pronunciation, audio examples, and interactive speech synthesis.

For languages like Bishnupriya Manipuri, where audio resources are limited, a dictionary-based TTS system offers an efficient way to make pronunciation accessible to learners and researchers.

Dictionary entry
      ↓
Pronunciation engine
      ↓
IPA
      ↓
Diphone synthesis
      ↓
Audio playback

2. Core Components of the System

A speech-enabled dictionary typically consists of several modules.

3. Dictionary Database Structure

A typical dictionary database may include fields such as:

When a user visits a word page, the system retrieves the corresponding record from the database.

4. Word Page Architecture

A dictionary word page usually performs the following tasks:

User searches word
       ↓
Server retrieves dictionary entry
       ↓
IPA pronunciation generated
       ↓
TTS button enabled
       ↓
User clicks “Play”
       ↓
Diphone synthesis
       ↓
Speech playback

This creates an interactive pronunciation experience.

5. Example Word Page

When the user presses the TTS button, the system calls the pronunciation API.

6. The Pronunciation API

A pronunciation API such as analyze_api.php acts as a bridge between the dictionary and the speech engine.

It receives a request containing a word ID or word string.

/analyze_api.php?id=4716

The API then returns pronunciation data in JSON format.

{
  "word": "দিশা",
  "ipa": "diʃa",
  "phonemes": ["d","i","ʃ","a"],
  "diphones": ["#-d","d-i","i-ʃ","ʃ-a","a-#"],
  "files": [
    "sil-d.wav",
    "d-i.wav",
    "i-sh.wav",
    "sh-a.wav",
    "a-sil.wav"
  ]
}

7. Audio Playback on the Word Page

Once the diphone file list is received, the browser loads and plays the corresponding audio files.

sil-d.wav
d-i.wav
i-sh.wav
sh-a.wav
a-sil.wav

The JavaScript engine plays them sequentially to synthesize the word.

8. Combining Recorded Audio and TTS

A dictionary may include both recorded word audio and synthetic TTS.

A common strategy is:

• If recorded word audio exists → play it
• If not → use diphone synthesis

if(word_audio_exists){
    play_recorded_audio();
}
else{
    play_diphone_tts();
}

This hybrid approach provides the best available pronunciation.

9. Linguistic Tools in the Dictionary

A speech-enabled dictionary can also include additional linguistic tools.

• IPA viewer
• phoneme trace panel
• diphone validator
• syllable breakdown
• audio alignment visualization

These tools transform the dictionary into a research platform.

10. Example Word Analysis Panel

ɔ kʰ ʃ ɔ r

#-ɔ
ɔ-kʰ
kʰ-ʃ
ʃ-ɔ
ɔ-r
r-#

Such panels help users understand the structure of pronunciation.

11. Benefits of Dictionary-Based TTS

Integrating TTS into a dictionary offers several advantages.

• users can hear pronunciation instantly
• language learners gain better phonetic awareness
• researchers can analyze phonological patterns
• audio resources grow gradually through dictionary expansion

This approach is particularly valuable for under-resourced languages.

12. Challenges

Some technical challenges must be addressed:

• missing diphone files
• inconsistent IPA conversion
• different pages using different rules
• audio normalization problems
• slow loading of multiple audio segments

These problems can be minimized through centralized pronunciation engines and validation tools.

13. Future Enhancements

A speech-enabled dictionary may evolve further by adding:

• sentence-level synthesis
• prosody modeling
• speech recognition
• automatic pronunciation learning
• neural TTS models

These improvements would transform the dictionary into a full language technology platform.

14. Conclusion

Integrating speech synthesis into an online Bishnupriya Manipuri dictionary creates a powerful educational and linguistic tool.

By combining lexical data, phonetic analysis, and diphone audio, the system allows users to hear accurate pronunciation directly from dictionary entries.

Such integration not only improves usability but also contributes to the preservation and documentation of the language.

Next Article

Article 10
Future Directions: Neural TTS and Advanced Speech Technology
for Bishnupriya Manipuri