Claims Analysis
Total Claims: 12
Claim 1
Independent Claim
75 words
A synthesis-based speech training apparatus, comprising: input means for entering an utterance; editing means for converting the utterance into a set of speech units and the onset and offset of each speech unit; synthesizer means for converting said set of speech units and onset and offset of speech units into a set of tongue-palate contact patterns arranged in a hierarchical order; and means for displaying the set of tongue-palate contact patterns for the entered utterance.
Claim 2
Dependent Claim
75 words
A synthesis-based speech training apparatus, comprising: input means for entering an utterance; editing means for converting the utterance into a set of speech units and the onset and offset of each speech unit; synthesizer means for converting said set of speech units and onset and offset of speech units into a set of tongue-palate contact patterns arranged in a hierarchical order; and means for displaying the set of tongue-palate contact patterns for the entered utterance.
Claim 3
Dependent Claim
26 words
The speech training apparatus of claim 1 wherein said synthesizer means provides the durations of the onset, offset, and contact steady-state portion of each speech unit.
Claim 4
Dependent Claim
25 words
The speech training apparatus of claim 2 wherein the times of onset, offset, and steady-state are assigned for each speech unit by said synthesizer means.
Claim 5
Dependent Claim
41 words
The speech timing apparatus of claim 3 wherein said synthesizer means stores the maximum area of tongue-palate contact for each speech unit along with at least one alternate contact area for each different context in which that speech unit is used.
Claim 6
Dependent Claim
24 words
The speech training apparatus of claim 4 wherein each contact area between the tongue and palate is defined by a set of 63 points.
Claim 7
Dependent Claim
63 words
The speech training apparatus of claim 1 further comprising means for measuring acoustic parameters of a spoken utterance, including measuring the tongue-palate contact pattern for the utterance; means for evaluating the similarity between the utterance being demonstrated and the measured acoustic parameters and measured tongue-palate contact pattern; and display means for displaying the differences between the utterance being demonstrated and the measured utterance.
Claim 8
Dependent Claim
63 words
A synthesis-based speech training method comprising the steps of: determining an utterance to be learned; converting said utterance into a speech unit and the onset and offset of the speech unit; converting said speech unit and onset and offset into a set of tongue-palate contacts; arranged in a hierarchical order; and displaying the set of tongue-palate patterns for the utterance to be learned.
Claim 9
Dependent Claim
63 words
A synthesis-based speech training method comprising the steps of: determining an utterance to be learned; converting said utterance into a speech unit and the onset and offset of the speech unit; converting said speech unit and onset and offset into a set of tongue-palate contacts; arranged in a hierarchical order; and displaying the set of tongue-palate patterns for the utterance to be learned.
Claim 10
Dependent Claim
36 words
The speech training method of claim 7 wherein said speech unit converting step comprises: storing an area of tongue-palate contact for each speech unit; and assigning times of onset, offset, and steady-state for each speech unit.
Claim 11
Dependent Claim
43 words
The speech training method of claim 8 wherein said storing an area of tongue-palate contact step comprises: storing the maximum area of tongue-palate contact along with at least one alternate contact area for each different context in which the speech unit is used.
Claim 12
Dependent Claim
28 words
The speech training method of claim 9 wherein said storing step comprises: storing each contact area between the tongue and palate by a specific set of 63 points.