Speech Synthesis by Rule: An Acoustic Domain Approach

Speech synthesis by rule is the method of converting from a discrete representation of speech in linguistic units, that is, phonemes and stress marks, to a continuous acoustic waveform. Fig. 1 shows the technique for carrying out this transformation. The figure shows t h a t the discrete input is converted to continuous control signals by the synthesis strategy. The synthesis strategy contains stored information about the phonemes and stored rules about the mutual effects of adjacent phonemes. The stored rules operate on the input sequence to produce the control signals for the synthesizer. The speech synthesizer converts the control signals to continuous speech. The synthesizer may be a terminal analog, a dynamic analog of the vocal tract, or a combination. 17 18 T H E BELL SYSTEM T E C H N I C A L J O U R N A L , JANUARY 1968 SPEECH SYNTHESIZER CONTINUOUS SPEECH CONTINUOUS CONTROL SIGNALS SYNTHESIS STRATEGY DISCRETE INPUT PHONEMES Fig. 1 -- Technique of Speech Synthesis.