Manual Speech Signal-to-Symbol Transformation

Part-1: Learn by pre-segmented examples

Step1 : Select an utterance of your choice from the list of examples provided.

Step2 : Note down the utterance in the native script of the language, and the word-level transliteration of the Indian language utterance using English alphabets.

Step3 : Study the subword-level (syllable-level and phoneme-level) transliterations by changing the subword unit.

Step4 : Now choose the subword unit as word, and note the list of words and their boundaries given in the table by the side of the speech waveform.

Step5 : Use the 'Zoom Slider', select and Play/Pause buttons to select a portion of the waveform, listen, and verify the boundaries of the subword units given in the table.

Step6 : Change the subword unit to syllable and repeat the experiment.

Step7 : Change the subword unit to phoneme and repeat the experiment.

Part-2: Segmentation with feedback

This part of the experiment is very similar to Part-1, except that the word/subword boundaries are not provided. The student is required to segment the given utterance into the chosen subword units by entering the correct subword boundaries.

Step1 : Select an utterance and choose the subword unit as word.

Step2 : Use the 'Zoom Slider', select and 'Play/Pause' buttons to select a portion of the waveform, listen, and enter the subword boundaries in the table.

Step3 : Look out for a message which pops up if the entered boundary value is 20ms more than the reference marking.

Step4 : Repeat the experiment by changing the subword unit to syllable.

Step5 : Repeat the experiment by changing the subword unit to phoneme.

Part-3: Segmentation without feedback

This part of the experiment is similar to the first two parts, but you have to choose your own utterance, transcribe and syllabify the utterance, record the utterance and segment it.

Step1 : Select a language of your choice and type in an utterance of your choice.

Step2 : Verify that the utterance you have typed is appearing correctly in the native script and press the done button.

Step3 : Now your chosen utterance and its constituent phonemes are displayed. Syllabify the utterance by grouping adjacent phonemes appropriately into syllables of the form ( C^VC^ ), where ( C ) denotes a consonant, ( V ) denotes a vowel, and ( ^* ) denotes zero or more occurrences of the consonant. Click on the 'Done' button.

Step4 : Now record the utterance by clicking on the 'Microphone: Start/Stop' button and then clicking on the 'Microphone: Start/Stop' button to stop recording. Load the utterance by clicking on the 'Play' button.

Step5 : Use the 'Zoom Slider', select and 'Play' buttons to select a portion of the waveform, listen, and enter the subword boundaries in the table.