200405194 玖、發明說明： 【發明所屬之技術領域】 厶匕、 月匕 < 可輸入 文字之可 本發明係關於一種具有聲音合成功 攜式終端裝置。 【先前技術】 現在之行動電話或PHS (註冊商標）終端 裝置除了基本之電話功能以夕卜，還可利用兩:可攜式終端 其收發功能或其他各種應用。例如也有時；表：：： = 網際網路連接功能等，㈣該等功能時亦需輸人以3 或URL (Uniform Resource Locator; _致資源定址器）等之文 字。當然於如PDA (可攜式資訊終端機）之可攜式終 置’ 一般亦可輸入文字。 利用如此之文字輸入功能之情況，例如於電子郵件或其 他文件之製作時，伴隨著文字之輸入，尤其是在行動電話^ 一般使用如圖10所示之鍵，由於為輸入手段之鍵（按鈕7之個 數限制，各文字與各键無法1對丨對應，伴隨按下特定次數 相同键等繁雜之輸入操作。例如輸入「早安」時，若是先 前之行動電話，要如「έ」鍵5次、「以」鍵1次、「^ 鍵3次、「岛」鍵3次般地進行許多鍵，來選擇輸入文字。 另—方面，鍵是否被接受，要按照其操作，以鍵單位發出 發音頻率不同之確認音或使鍵本身發光，藉由此等可以確 認。 [本發明欲解決之問題] 然而，如上所述，於键與欲輸入之文字未對1對應之行動 電話等之可攜式終端裝置，僅有键單位之確認音等於該時 83900 200405194 =法確認選擇之文字（輸入候選文字），為了正確進行文 子輻入，要以目視確認按照键操作所顯示之輸入候選文、 來確定輸入所希望之輸入文字。另一方面，不藉由 依賴按鈕操作時之記憶之情況’如前所伴隨繁雜之輸入操 作，就會在錯誤輸入狀態下繼續輸入，結木 此外，使用者為視障者時之文字輸入，明顯=為再= 難。 本發明係鑒於上述之點而成者，於可輸入文字之可攜式 終端裝置，提供一種藉由合成之聲音可確認輸入候選 之可攜式終端裝置。 【發明内容】 為解決上述之課題，申請專利範圍第1項之發明，其特徵 在於：係一種可以輸入文字之可攜式終端裝置，其包含操 作手段，其係進行為了輸入文字之特定操作者；顯示手段， 其係顯示與該操作相對應之輸入候選文字者；及輸入控制 手段’係藉由由該操作手段接受確定輸入文字之操作，將 於員示於遠頭示手段之輸入候選文字做為輸入文字輸入者； 且具備聲骨合成手段，其顯示該輸入候選文字於該顯示手 段時’將該輸入候選文字之發音聲音合成後輸出。 此外’申請專利範圍第2項之發明，係於申請專利範圍第 1項之可攜式終端裝置，前述可攜式終端裝置係行動電話； 且將前述聲音合成手段使用於聲音合成之音源與使用於前 述行動電話所具備之來電音生成之音源共用。 再者，申請專利範圍第3項之發明，係於申請專利範圍第 83900 200405194 2项^可攜式終端裝置，前述聲音合成手段使用於聲音合成 之首源係FM骨源或波形表（WT)音源。 申叫專利範圍第4項之發明，係於申請專利範園第丨項之 可攜式終端裝置，前述操作手段包含可手動操作之複數按 紐’於各按紐分擔可輸入之複數文字。 申請專利範圍第5項之發明，係於申請專利範圍第丨項之 可攜式終端裝置，前述操作手段可指示該輸入候選文字之 輸入確定，前述顯示手段顯示指示該輸入確定之文字。 申請專利範.圍第6項之發明，係輸人文字於可攜式終端裝 置《方法’其進行：操作程序，其係進行為了輸人文字之 顯示程序，其顯示與該操作相對應之輸入候選文字； 聲骨合成程序’其顯示該輸人候選文字時，將該輸入候選 又字之發音聲音合成後輸出；及輸人程序，其藉由由該操 作程序接受確定輸人文字之操作，將該被顯示且被發音之 幸细入候選文字做為輸入文字輸入。 申請專利範圍第7項之發明，係㈣可攜式終端裝置被執 行之程式’該可攜式終端裝置具備為了對可攜式終端裝置 輸入又字而所操作之操作器；其進行：顯示程序，其顯示 由該操作所指定之輸入候選文字；聲音合成程序，立顧亍 該輸入候選文字時，將該輸人候選文字之發音聲音合成後 輸出；及輸入程序’其藉由由該操作器接受確定輸入文字 之操作，將該被顯示且被發音之輸人候選文字做為輸入文 字輸入。 於本發明之可攜式終端裝置 顯示與為了文字輸入之特 83900 200405194 疋铵作 < 輸入候選文字時，聲音合成手段將該輸入候選文 字义發首聲首合成後輸出。藉此，使用該可攜式終端裝置， 例如仃動電話之使用者，藉由聽聲音合成之輸入候選文字 ，發音，可以確認該輸入候選文字，故無需如先前目視確 〜知入候選文字，方便性提高。此外，即使使用者為視障 者文字輸入也答易。再者，行動電話之情形，藉由將聲 首合成手段用於聲音合成之音源與用於前述行動電話所具 t之來包曰生成之背源共用，故無需為了聲音合成手段而 追加新裝置，可以抑制製造成本之增加。 【實施方式】 、以下參照圖面說明本發明之實施型態。此外，於以下之 4明中，對同一構成要素賦予相同之符號。 —於圖1顯示本發明之可攜式終端裝置一實施型態之行動 包治 < 結構。於圖i中，符號“係cpu (中央處理裝置），藉 由執行下逑之各種控制程式來控制行動電話丨之各部動 苻唬lb係ROM (Read 〇niy Mem〇ry ;唯讀記憶體）。此r〇m =儲存進行CPU la執行之傳送、來電等控制之各種電話功 匕秸式私子郵件之作成或控制其收發之郵件收發功能程 ^辅助樂曲播放處理之矛呈式、幸甫助聲音合成處理之程式 等又私式、或預先記錄之樂曲資料及伴奏資料、聲音合成 :斤必叙參數或相關資訊等資料。該程式被設計成進行： :π程序’其顯示由該操作所指定之輸入候選文字；聲音 Β成私序其頭示该輸入候選文字時，將該輸入候選文字 、X曰耳㈢5成後輸出；輸入程序，其藉由由該操作器接 486 83900 200405194 受確足輸入文字之操作，將該被顯示且被發音之輸入候選 文字做為輸入文字輸入。 付唬lc係RAM (Random Access Memory ;隨機存取記憶 體），係設SCPU la之工作區域、下載之樂曲資料或伴奏資 料之儲存區域及儲存接收之電子郵件資料之郵件資料儲存 區域等。符號Id係通信裝置’進行以天線_收之信號解 調’並將傳送之信號調變而提供給天線n。此外，符號^ 係輸入裝置，具有手動操作手段，其由包含設置於行動電 話1本體之「〇」〜「9」之撥號按鈕之各種按紐（鍵；後述圖 1之叫構成；檢測由此等手動操作手段之輸入。此操作手段 用於進行為了文字輸人之特定操作，包含可手動操作之複 數按奴，於各按叙預先分配可輸入之複數文字。 符號U係通話裝置，以通訊裝置_調之受話信號由此通 活裝置lf所具備之聲音⑵㈣所解碼後，由同裝置 D/A轉換器（皆未圖示）所D/A轉換後由受話口（耳揚聲 輸出。另-方面，由送話口 (麥克風）lh輸入之聲音二g 同裝置所具備之A/D轉換器所數位 化 ,^ ^ ^ U ?水^冋裝置所具備 Μ首CODEC (皆未圖示)所壓縮編碼後，由通信、 基地台傳送。作為此通話裝置lf之編碼/解碼° ^ 厂一—;代碼激發線性預測編：)方= ADPCM (適應差分PCM編碼）方式等之聲立资 ^二或 縮編碼/解碼方式。 曰”，炙向效率壓 符號li係音源裝置，播放選擇之樂曲資 或保留音，由背面揚声先哭 、 乍為來電音 “面揚…輸出。此外，電子郵件作成時 4ft： 83900 200405194 等進行文字輸入時，接受CPU la之控制，將其輸入候選文 字聲音合成，將合成之聲音由背面揚聲器lj輸出。有關此 聲音合成之詳細後述。此外，符號lk係顯示裝置，由LCD (Liquid Crystal Display ;液晶顯示器）所構成，進行與電話功 能或電子郵件收發功能之項目單或撥號按鈕等各種按鈕之 操作相對應之顯示。於文字輸入時，顯示輸入候選文字或 所確定之輸入文字。再者，各功能塊經由滙流排10進行資 料或命令之授受。 圖11係顯示本發明一實施型態之行動電話之外觀形狀之 圖。如圖所示，本行動電話係小型構造，其一體裝入操作 部，其基本上進行有關去電及來電之操作；通話部，其按 照該操作使通話可能；及顯示部，其可以顯示有關至少操 作之資訊。具體而言，如圖示一般，行動電話1具備無線收 發用之天線11、受話器（揚聲器）lg及送話器（麥克風）lh， 同時具有包含撥號键等操作键之輸入手段le及圖像顯示裝 置（顯示器）lk。此行動電話1可以將個人名及電話號碼等之 電話簿資訊顯示於顯示器lk。亦可以將接收之電子郵件顯 示於顯示器lk。並可以依據本發明之方式輸入文字。 於此，說明有關音源裝置li之詳細。於本實施型態，原 樣地利用使用於來電音等生成之習知音源裝置，實現輸入 候選文字之發音之聲音合成。於圖2顯示音源裝置li之概略 結構。 於圖2中，符號21之輸出入I/F (介面）經由滙流排10,由CPU la接受為了播放來電旋律等音樂之樂曲序列資料或命令， -10- 83900 200405194 並且係為了將下述FIFO 22之狀態通知輸出於CPU la之介面 電路。FIFO 22係包含 FIFO記憶體（First In First Out memory ; 先進先出記憶體）之電路，暫時保持被給之樂曲序列資料 (①），提供給顯示於順次符號23之排序器（②）。此外，FIFO 22 通知CPU la記憶體之空間狀況（⑤），記憶體變空（Empty)之前 接受繼續之樂曲序列資料之轉送。 排序器23由CPU la接受發音開始/發音結束等命令（⑥），於 開始發音時，解釋由FIFO 22接受之樂曲序列資料之同時， 計量時機而將.各種參數或控制信號提供給FM音源24 (詳細 後述）或WT音源25 (③、④），驅動該等音源。WT音源25如眾 所皆知，係將各種樂器聲或聲音等數位錄音，藉由大概或 反覆讀出預先儲存之波形記憶體2.6之波形資料，化實地再 現原本之樂器聲或聲音等。 FM音源24及WT (波形表）音源25之輸出以加法器27加 算，其輸出於數位/類比轉換器（未圖示）轉換為類比資料， 供給背面揚聲器1 j (圖1)。一般於音源裝置1 i，各音源係 經由FIFO 22及排序器23驅動，惟被要求即時性（即時響應性） 之效果音之種類，CPU la不經由FIFO 22及排序器23，而直 接驅動FM音源24或WT音源25。於本實施型態中之聲音合成 亦同樣地，CPU la直接驅動各音源。再者，波形記憶體26 係使用ROM構成。 其次說明有關FM音源24。FM音源24—般組合複數圖3所示 之運异器3 0及加法器所構成。如圖3所示，1個運算器3 〇包 含：SIN波形表31，其記憶於sin波形（正弦波之波形）之各相 4δ9 83900 -11 - 200405194 位角點之波形振幅值；相位產生器（pG) 32 ,其由排序器 或CPU la接受頻率參數，基於此頻率參數生成為了控制由 SIN波形表31使其輸出之SIN波形資料之頻率及相位之相位 位址信號後輸出加法器33，其將輸入信號及上述相位位址 加起來後提供給SIN波形表31 ;色絡產生器（EG) 34，其由排 序器23或CPU la接受振幅參數，生成為了控制由該運算器3〇 輸出之波形之振幅之色絡信號（振幅係數）後輸出，及乘法 器3S，其乘以SIN波形表31之輸出及色絡產生器（EG)%之輸 出。 於如此構成之運算器30,記憶於SIN波形表31<SIN波形之 振幅值係依據包含經由加法器33所提供之相位位址信號之 信號依序被讀出。因此，於此運算器3〇，藉由使讀出記憶 於SIN波形表31之波形振幅值之速度變化，即藉由適當控制 提供於SIN波形表31之相位位址信號，可以改變音高。例 如，放f更項出速度則可生成低的音’提高讀出速度則可生 成咼的首。再者，相位產生器（PG) 32接受重設信號，則將 輸出之相位位址信號重設（使由SIN波形表31讀出之位址回 到初始值）。 FM首源24如圖4(a)所示，將如此之運算器3〇複數串級連 接，或如同圖（b)所示，再進一步使用加法器，加上運算器 30之輸出’以各式各樣地組合複數運算器3〇及加法器，可 以生成無限多樣種類之聲音。於本實施型態中，利用揭示 於特公昭58-53351號公報等之所謂CSM聲音合成之技術，使 用具備於此行動電話ltFM音源24實現於該行動電話丨之聲 83900 -12- 首合成 ο 於此，說明有關Ρ、+ 短時間内可視為=聲音合成之原理。-般聲音於 間内聲音之頻譜視為―：：此，CSM聲音合成係在短時 將數礙數十ms之短時門而t行聲音之合成。具體而言， 之和來表現聲音。依據了首視為穩定，以數個正弦波 表示成： 離政時間表現，聲音之時間系列{xt} xt sinc〇i t + 4- Λ · .·· + AnSl⑽nt -中t係表示離散時刻敕 為4〜6個左右），〇〇.係第iiE 係正弦波成分之個數（一般 T r Λ 1係弟1正弦波成分之角頻率（Ο^θπ)，弋係 正弦波成分之振幅。 於此CSM聲音合成，祖' 〇。成對相上述⑴式表示之模式，給予 :數{ο^.,.ωη八…人} ’藉由⑴式，就各時刻t求出合成聲 =系列{M。此時，對於有聲音（母音或濁子音等），因有聲 首具有周期性，故每此週期（間隔週期）將於⑴式之時刻t重 設成零而使相位初始化，另一方面，對於無聲音，因不具 周期性，故給丁隨機週期，即於隨機之週期冑時刻t重設而 使相位隨機地初始化。&此合成之聲音信號之時間系列接 近於人之聲音。 其次，就明有關此CSM聲音合成技術適用於FM音源24 (參 照圖5)。於（1)式所表示之各正弦波之成分可以使用前述之 運算器30生成。即’藉由對應於各正弦波之成分之SIN波形 表31 ’於時間系列使正弦波輸出（此時，各運算器％之輸入 信號定為零，相位產生器（PG) 32提供為了由SIN波形表31讀 83900 -13 - 200405194 出正弦波之波形資料之相位位址信號（位址）），藉由下一級 之乘法器35，使其具有由色絡產生器（EG) 34所提供之振 幅，可以由各運算器30得到（1)式之各正弦波成分之信號之 輸出。然後，藉由以加法器50加上此等之輸出，可得到合 成聲音訊號之系列{xt}。於CSM聲音合成，對於有聲音，每 其週期將時刻t重設成零，使相位初始化，並對於無聲音， 於隨機週期將時刻點t重設成零，使相位初始化，此相位之 初始化可藉由對於相位產生器（PG) 32於各週期提供重設信 號，使相位初始化來進行。 如以上，於使用FM音源24之CSM聲音合成，藉由合成複 數共振峰音，其係由提供給相位產生器（PG) 32之頻率參數 或重設信號及提供給色絡產生器（EG) 34之振幅參數之3要 素所合成者，決定音素，可進行聲音合成。例如聲音合成 「櫻花」時，藉由每數mS至數十mS設定複數組之上述3要 素，合/S/4/A/4/K/->/U/4/R/->/A/之6音素後發音。再者，小 的「〇」、「爷」等或英文字之小寫等提高音程等來區別， 有關其他之記號亦預先決定易了解之說法使其發音即可。 提供給各運算器30之上述3要素，每個音素預先定義，登 記於ROM lb。此外，有關構成各文字之各音素之資訊，例 如「$」之情形，該文字由音素/S/—/^/所構成等之資訊， 亦同樣地登記於ROM lb。行動電話1於文字輸入時（後述之 文字輸入模式時），與先前相同地顯示對應於鍵操作之輸入 候選文字。然後，再進一步於顯示該輸入候選文字時，參 照有關登記於ROM lb之構成該輸入候選文字之音素之資 83900 -14- 200405194 訊，由所得之資訊再參照上述3要素之參數，其對應於構成 為如入候選文字之晋素，每數mS至數+mS，提供頻率參數 或重設信號給相位產生器（PG) 32，並提供振幅參數給色絡 產生器（EG) 34，將輸入候選文字之發音聲音合成後輸出。 此外，於本發明之型態，使用FM音源24實行CSM聲音合 成，惟當然顯然使用WT音源25亦可聲音合成。例如聲音八 成「樓花」時，「$」、「<」、「；」數位錄音後料 體，撥放此等即可。然而，使用跟音源Μ進行⑽ 戽曰口成，必要之參數（資料）少即可，較為有利。 、其次:料如此構成之本實施型態之可攜式終端裝置工 =電寺待模式時之動作，參照圖6所示之動作流程圖加以 :二於此來電等待模式，音源裝置U係為播放來電旋律 CPU。判斷有無來電(步驟S61)，反覆此判斷直至 有來電為止（被判定立σ & 步驟如之判斷判定為^°於此’作為有來電。則教 J疋為疋，移至步驟S62。200405194 发明 Description of the invention: [Technical field to which the invention belongs] The dagger, the moon dagger < The input of text is possible The present invention relates to a portable terminal device with sound integration. [Prior art] Existing mobile phones or PHS (registered trademark) terminal devices, in addition to basic telephone functions, can also use two: portable terminals, their transceiver functions, or various other applications. For example, sometimes; Table ::: = Internet connection function, etc. When these functions are entered, the text such as 3 or URL (Uniform Resource Locator; Of course, in a portable terminal such as a PDA (Portable Information Terminal), you can also enter text. In the case of using such a text input function, for example, in the production of an email or other document, accompanied by the input of text, especially on a mobile phone ^ Generally, the key shown in FIG. 10 is used because it is a key for input means (button The number of 7 is limited, and each character cannot correspond to each key. It is accompanied by complicated input operations such as pressing the same key a certain number of times. For example, when entering "Good Morning", if it is a previous mobile phone, it must be the same as the "Hand" key Press the "5" key once, the "^" key 3 times, and the "Island" key 3 times to select the input text. On the other hand, whether the key is accepted or not, follow the operation in key units. A confirmation sound with a different pronunciation frequency is emitted or the key itself is illuminated, so that the key can be confirmed. [Problems to be Solved by the Present Invention] However, as described above, in a mobile phone or the like where the key and the text to be input do not correspond to 1. For portable terminal devices, only the confirmation sound of the key unit is equal to the time 83900 200405194 = method to confirm the selected text (enter the candidate text). In order to correctly enter the text, you must visually confirm and press the key. The displayed input candidate text is used to determine the desired input text. On the other hand, without relying on the memory of the button operation, as in the previous complicated input operation, input will continue in the wrong input state In addition, the text input when the user is visually impaired is obviously = difficult to re- = difficult. The present invention is made in view of the above points, and provides a method for synthesizing a portable terminal device capable of inputting text. A portable terminal device that can confirm input candidates by voice. [Summary of the Invention] In order to solve the above-mentioned problem, the invention in the scope of patent application No. 1 is characterized in that it is a portable terminal device that can input text, which includes operations Means for performing a specific operator for inputting text; display means for displaying a candidate for inputting text corresponding to the operation; and input control means' receiving an operation to determine the input text by the operation means, The input candidate text displayed in the remote head display means is used as the input text input person; and the sound bone synthesis means is provided, which displays the input When the input candidate text is in the display means, the pronunciation and sound of the input candidate text are synthesized and output. In addition, the invention in the second patent application scope is a portable terminal device in the first patent scope application. The terminal device is a mobile phone; and the sound source used for the sound synthesis by the aforementioned sound synthesizing means is shared with the sound source used for the generation of the incoming call tone provided by the aforementioned mobile phone. Furthermore, the invention in the third scope of the patent application is for the application Patent scope No. 83900 200405194 2 ^ Portable terminal devices, the aforementioned sound synthesis means is used for the first source of sound synthesis is FM bone source or waveform table (WT) sound source. The invention claimed in item 4 of the patent scope belongs to the application In the portable terminal device of the patent item No. 丨, the aforementioned operation means includes a plurality of buttons that can be manually operated, and the plural characters that can be inputted are shared among the buttons. The invention in the fifth scope of the patent application is a portable terminal device in the first scope of the patent application. The aforementioned operation means may indicate the input determination of the input candidate text, and the aforementioned display means displays the text indicating the input determination. The invention claimed in item 6 of the patent application is for inputting characters in the portable terminal device "Method" which is performed: operation procedure, which is a display procedure for inputting characters, which displays input corresponding to the operation Candidate text; a phonological synthesizing program 'that displays the input candidate text and synthesizes the pronunciation sound of the input candidate when it is displayed; and an input program that accepts the operation to determine the input text by accepting the operation program, The displayed and pronounced fortunately entered candidate text is used as input text input. The invention in the seventh scope of the patent application is a program executed by the portable terminal device. The portable terminal device is provided with an operator which is operated to input a word to the portable terminal device; , Which displays the input candidate text specified by the operation; a sound synthesis program that, when considering the input candidate text, synthesizes the pronunciation of the input candidate text and outputs the output sound; and the input program 'which uses the operator Accept the operation of determining the input text, and input the displayed and pronounced input candidate text as input text. When the portable terminal device of the present invention displays a special feature for text input 83900 200405194 osmium as < input candidate text, the voice synthesis means first synthesizes the input candidate text and first synthesizes it and outputs it. With this, a user using the portable terminal device, such as a mobile phone, can confirm the input candidate text by listening to the input candidate text synthesized by the voice, and pronounce it, so there is no need to confirm the candidate text visually as previously, Convenience is improved. In addition, even if the user inputs text for the visually impaired, it is easy to answer. Furthermore, in the case of a mobile phone, a sound source using the sound-synthesizing method for sound synthesis is shared with a background source used for the aforementioned mobile phone, so there is no need to add a new device for the sound synthesis method. , Can suppress the increase in manufacturing costs. [Embodiment] An embodiment of the present invention will be described below with reference to the drawings. In addition, in the following 4 explanations, the same reference numerals are assigned to the same constituent elements. -Figure 1 shows an action of an implementation of the portable terminal device of the present invention. Package < structure. In Figure i, the symbol "is a CPU (Central Processing Unit) and controls various parts of the mobile phone by executing various control programs below". The lb series ROM (Read 〇niy Mem〇ry; read-only memory) . This r0m = stores the various telephone functions to control the transmission and incoming calls performed by the CPU la, or creates and sends and receives the mail receiving and sending functions. ^ A spear presentation that assists music playback processing. Xingfu Programs that assist in sound synthesis processing, such as private or pre-recorded music data and accompaniment data, sound synthesis: necessary parameters or related information, etc. The program is designed to perform:: π program 'whose display is performed by this operation The designated input candidate text; when the voice B is in the private sequence and its head indicates the input candidate text, the input candidate text and X-Year ears are output after 50%; the input program is received by the operator through 486 83900 200405194. The operation of inputting text is sufficient, and the input candidate text that is displayed and pronounced is used as input text input. The lc is RAM (Random Access Memory), which is set by SCPU la. As the storage area, downloaded music data or accompaniment data storage area, and received e-mail data storage area, etc. The symbol Id is the communication device 'demodulates with the antenna_received signal' and modifies the transmitted signal. It is provided to the antenna n. In addition, the symbol ^ is an input device with a manual operation means, which includes various buttons (keys including dial buttons "0" to "9" provided on the body of the mobile phone 1; Call the composition; detect the input by such manual operation means. This operation means is used to perform specific operations for text input, including a plurality of manual buttons that can be manually operated, and a plurality of input characters that can be input are pre-assigned in each column. Symbol U It is a communication device. It is decoded by the communication device _ tuned by the voice signal of the communication device lf, and then converted by the D / A converter of the same device D / A (not shown). (Speaker output. In the other aspect, the sound input from the speaker port (microphone) lh 2g is digitalized by the A / D converter provided in the device. ^ ^ ^ U? COD After being compressed and coded by EC (neither shown), it is transmitted by the communication and base station. As the encoding / decoding of the communication device lf ° ^ Factory 1 —; Code Excited Linear Prediction Code :) Square = ADPCM (adaptive differential PCM coding) The sound of the method is based on the second or reduced encoding / decoding method. Said ", to the efficiency pressure symbol li series sound source device, play the selected music material or reserved sound, cry from the back, first cry, at first the call sound" face Yang ... output. In addition, when the e-mail is created, 4ft: 83900 200405194, etc. When text input is performed, it is controlled by the CPU la, and the input candidate text is synthesized, and the synthesized sound is output from the rear speaker lj. The details of this sound synthesis will be described later. In addition, the symbol lk is a display device composed of an LCD (Liquid Crystal Display; liquid crystal display), and displays corresponding to operations of various buttons such as a menu of a telephone function or an e-mail transmission and reception function, or a dial button. During text input, input candidate text or determined input text is displayed. In addition, each functional block transmits or receives data or commands via the bus 10. Fig. 11 is a diagram showing the appearance of a mobile phone according to an embodiment of the present invention. As shown in the figure, the mobile phone has a small structure, which is integrated into an operation unit, which basically performs operations related to outgoing and incoming calls; a call unit, which makes a call possible according to the operation; and a display unit, which can display relevant information. Information on at least operations. Specifically, as shown in the figure, the mobile phone 1 is provided with an antenna 11 for wireless transmission and reception, a receiver (speaker) lg, and a microphone (microphone) lh, and an input means le including an operation key such as a dial key and an image display. Device (display) lk. This mobile phone 1 can display phonebook information such as a personal name and a phone number on the display lk. The received e-mail can also be displayed on the display lk. And can input text according to the method of the present invention. Here, the details of the sound source device li will be described. In this embodiment, a conventional sound source device used for generating a ringing tone or the like is used as it is to realize sound synthesis of inputting pronunciation of candidate characters. A schematic structure of the sound source device li is shown in FIG. 2. In Figure 2, the input / output I / F (interface) of symbol 21 is received by the CPU la via the bus 10 to receive sequence data or commands for playing incoming melody and other music. -10- 83900 200405194 The status notification of 22 is output to the interface circuit of the CPU la. FIFO 22 is a circuit that includes a FIFO memory (First In First Out memory), which temporarily holds the sequence data of the given song (①), and provides it to the sequencer (②) displayed in the sequential symbol 23. In addition, the FIFO 22 notifies the CPU of the memory space condition (⑤), and it continues to transfer the music sequence data before the memory becomes empty (Empty). The sequencer 23 accepts commands such as start of speech / end of speech (⑥) by the CPU la. At the beginning of the speech, it interprets the sequence data of the music accepted by the FIFO 22, and measures the timing to provide various parameters or control signals to the FM sound source 24. (Described in detail later) or WT source 25 (③, ④) to drive these sources. As is well known, the WT sound source 25 is a digital recording of various instrument sounds or sounds, and the waveform data of the waveform memory 2.6 stored in the waveform memory 2.6 is read out roughly or repeatedly to reproduce the original instrument sound or sound. The output of the FM sound source 24 and the WT (waveform table) sound source 25 are added by the adder 27, and the output is converted into analog data by a digital / analog converter (not shown) and supplied to the rear speaker 1 j (Figure 1). Generally, in the sound source device 1 i, each sound source is driven by the FIFO 22 and the sequencer 23, but the type of effect sound that is required to be immediate (immediate response). The CPU la directly drives the FM without passing through the FIFO 22 and the sequencer 23. Sound source 24 or WT sound source 25. In the same manner as the sound synthesis in this embodiment, the CPU la directly drives each sound source. The waveform memory 26 is configured using a ROM. The FM sound source 24 will be described next. The FM sound source 24 is generally composed of a complex number 30 and an adder shown in FIG. As shown in FIG. 3, one computing unit 3 includes: a SIN waveform table 31, which is stored in each phase of a sin waveform (sine wave waveform) 4δ9 83900 -11-200405194 waveform corner value; the phase generator (PG) 32, which receives a frequency parameter by the sequencer or CPU 1a, generates a phase address signal for controlling the frequency and phase of the SIN waveform data output by the SIN waveform table 31 based on the frequency parameter, and outputs the adder 33, It adds the input signal and the above-mentioned phase address to the SIN waveform table 31; the color network generator (EG) 34, which receives the amplitude parameter by the sequencer 23 or the CPU 1a, and generates the output to be controlled by the processor 30. The amplitude of the color waveform signal (amplitude coefficient) of the waveform is output, and the multiplier 3S is multiplied by the output of the SIN waveform table 31 and the output of the color generator (EG)%. In the thus constructed arithmetic unit 30, the amplitude values stored in the SIN waveform table 31 < SIN waveform are sequentially read out based on the signals including the phase address signals provided through the adder 33. Therefore, in this arithmetic unit 30, the pitch can be changed by changing the speed of reading the waveform amplitude value stored in the SIN waveform table 31, that is, by appropriately controlling the phase address signal provided in the SIN waveform table 31. For example, if you release f more, you can generate a lower sound ', and if you increase the read speed, you can generate a fret. Furthermore, when the phase generator (PG) 32 receives the reset signal, it resets the output phase address signal (returns the address read from the SIN waveform table 31 to the initial value). As shown in Fig. 4 (a), the FM first source 24 connects such an arithmetic unit 30 complex cascade, or as shown in Fig. (B), further uses an adder and adds the output of the arithmetic unit 30 to each Combining the complex arithmetic unit 30 and the adder in various ways can generate an infinite variety of sounds. In this embodiment, the so-called CSM sound synthesis technology disclosed in Japanese Patent Publication No. 58-53351 is used to implement the sound of the mobile phone ltFM sound source 24 provided in the mobile phone 丨 voice 83900 -12- the first synthesis ο Here, the principle that P, + can be regarded as = sound synthesis in a short time is explained. -The spectrum of ordinary sound in the room is regarded as ::: Therefore, CSM sound synthesis is the synthesis of t-line sounds that interferes with short-term gates of several tens ms in a short time. Specifically, the sum represents the sound. Based on the first view of stability, it is represented by several sine waves as: the time of departure, the time series of sounds {xt} xt sinc〇it + 4- Λ ·. ·· + AnSl⑽nt-where t is the discrete time 敕 is About 4 to 6), 〇〇. Is the number of iiE sine wave components (generally T r Λ 1 is the angular frequency of the sine wave component (0 ^ θπ), 弋 is the amplitude of the sine wave component. This CSM sound synthesis, the ancestor '〇. Pairwise phase of the above-mentioned expression, given: the number {ο ^.,. Ωη 八… 人}' through the expression, to find the synthesized sound at each time t = series { M. At this time, for sounds (vowels, voiced consonants, etc.), since the voiced head has periodicity, each period (interval period) will be reset to zero at the time t of the formula to initialize the phase, on the other hand For non-sound, because there is no periodicity, the random period is given, that is, the phase is randomly initialized at a random period 胄 time t reset. &Amp; The time series of this synthesized sound signal is close to human voice. Second It is clear that this CSM sound synthesis technology is applicable to FM sound source 24 (refer to Figure 5). (1) The indicated components of each sine wave can be generated using the aforementioned arithmetic unit 30. That is, 'the SIN waveform table 31 corresponding to the components of each sine wave is used to output a sine wave in the time series (at this time,% of each operator The input signal is set to zero. The phase generator (PG) 32 provides the phase address signal (address) for reading the sine wave waveform data from the SIN waveform table 31 83900 -13-200405194. 35, so that it has the amplitude provided by the color network generator (EG) 34, and the output of the signal of each sine wave component of formula (1) can be obtained by each processor 30. Then, by adding 50 These outputs can be used to synthesize a series of {xt} sound signals. For CSM sound synthesis, for sound, reset the time t to zero every cycle to initialize the phase, and for no sound, set the time at random cycles. The point t is reset to zero to initialize the phase. The initialization of this phase can be performed by providing a reset signal to the phase generator (PG) 32 in each cycle to initialize the phase. As above, the CSM of the FM source 24 is used. sound By synthesizing the complex formant tone, it is a combination of the three parameters of the frequency parameter or reset signal provided to the phase generator (PG) 32 and the amplitude parameter provided to the color generator (EG) 34, Determining the phoneme can be used for voice synthesis. For example, when the voice synthesis is "Sakura", the above three elements of the complex array are set by every several mS to several tens of mS, which is / S / 4 / A / 4 / K /-> / U / 4 / R /-> / A / after 6 phonemes are pronounced. In addition, the small "〇", "lord", etc., or the lower case of the English word, etc. are used to increase the interval, etc., and other symbols are also determined in advance. Know what it means to pronounce it. The above-mentioned three elements provided to each computing unit 30 are defined in advance for each phoneme and registered in the ROM lb. In addition, information about each phoneme constituting each character, such as "$", and the information that the character consists of the phoneme / S /-/ ^ / are also registered in the ROM lb in the same manner. When the mobile phone 1 is inputting a character (in a character input mode described later), the input candidate characters corresponding to the key operation are displayed as before. Then, when displaying the input candidate text, refer to the information on the phonemes constituting the input candidate text registered in ROM lb 83900 -14- 200405194. From the obtained information, refer to the parameters of the above 3 elements, which corresponds to It is constituted as a prime element that enters the candidate text, and provides a frequency parameter or a reset signal to the phase generator (PG) 32, and an amplitude parameter to the color generator (EG) 34 for every mS to + mS. The pronunciation of the candidate text is synthesized and output. In addition, in the form of the present invention, CSM sound synthesis is performed using the FM sound source 24, but of course it is also possible to use the WT sound source 25 for sound synthesis. For example, when the sound is "off-plan", "$", "<", and ";" are digitally recorded materials. However, it is more advantageous to use the audio source M to perform vocalization, and only a few necessary parameters (data) can be used. 2.Second: The portable terminal device of this implementation type is expected to constitute the action when it works in the electric standby mode, referring to the operation flowchart shown in Figure 6: Second, in this call waiting mode, the sound source device U is Play incoming melody CPU. It is judged whether there is an incoming call (step S61), and the judgment is repeated until there is an incoming call (it is judged that σ & step is judged as ^ ° here 'as an incoming call. Then teach J 疋 as 疋, and go to step S62.
將預先選擇設定作為來…… 万、八S62CPUL 至音源裝置U。於音源裝曲之樂曲序列資料轉送 成來電旋律，括缽 1土万；收到 <樂曲序列資料，合 成不％捉律持續播放該來電旋律。 其次，判斷通怒鍵於開狀態或 譯判定通話鍵為關狀態時(被判定卿 (判定為是時)回到=步斷:梅路斷路時 回到步驟S63。 7路_路時（判定為否時） 83900 -15- 493 200405194 另一方面，於步驟S63判定通話鍵為開狀態時（被判 是時），CPU la給予音源裝置li終止來電旋律播放之命令（+ 驟S65)。於此階段，音源裝置Π終止目前正在播放之來電旋 律之播放。然後’進行-般通話時之處理(步驟s66)。於接 下來之步驟S67判斷結束通話鍵係開狀態或關狀態，反覆此 判斷直到結束通話鍵成為開狀態為止（被判定為1為止）。 然後，於此判斷，判定結束通話鍵為開狀態時（判=為是時） 移至步驟S68。然後，於步驟S68進行結束通話時之處理（線 路斷路），回到步驟S61。以上說明了於來電等待模式之由 來電至線路斷路之動作。 其次，對於如此構成之本實施型態之可攜式終端裝置i 之文字輸入（文字輸入模式）時之動作，參照圖？所示之動作 流程圖加以說明。於此係經由使用者特定之操作，該可攜 式終端裝置1處於文字輸入模式者。此外，於以下及圖7了 “NKN”（新鍵號碼）及“OKN”（舊鍵號碼）為變數，“ —，，係表示 游標輸送鍵。再者，於OKN作為初始值設定數值键之代碼 ^外之代碼。此外，變為文字輸入模式的係例如電子郵件 或時程表或其他文件之作成時，或網際網路連接時之url 义輸入時等，文字輸入成為必要時變成此模式。 首先，於步驟S71判斷鍵是否被碰觸（鍵開）。然後，反覆 此判斷直到使用者碰觸鍵為止。於此，輸入裝置le檢測使 用者之鍵碰觸，檢測出鍵碰觸時，將表示鍵被碰觸之鍵號 碼通知給CPU la。CPU la從輸入裝置k到收到鍵號碼通知， 判疋為沒有鍵碰觸。於此係檢測出鍵碰觸（於步驟S71，是 83900 -16- 200405194 的判定）。此時，CPU la由輸入裝置le接到键號碼通知，將 接到之键號碼設定於變數NKN (步驟S72)。 其次，判斷設定於變數NKN之代碼（於此為鍵號碼）是否為 數值键之代碼（步騾S73)。於此，判定被設定於變數NKN之 代碼非數值键之代碼，則移至步騾S74。於步騾S74，再判 斷被設定於變數NKN之代碼是否為游標輸送鍵（「4」）之代 碼。於此步驟S74之判斷，判定被設定於變數NKN之代碼非 游標輸送键之代碼（否之判定時），則執行對應於另外規定 之其他鍵之處理（步騾S75)。然後，移至步騾S76。 於步騾S76，將變數NKN之代碼設定於變數OKN後，回到 步騾S71。此外，使用者之键操作為特定之模式變更操作 時，即被設定於變數NKN之代碼為對應於此模式變更操作 之鍵之代碼時，於步驟S75脫離圖7之文字輸入模式之流 程，終止文字輸入模式。 另一方面，於步騾S74之判斷，判定被設定於變數NKN之 代碼為游標輸送鍵之代碼（是之判定時），移至下個步騾 S77。然後，於步騾S77再判斷被設定於變數OKN之代碼是否 為數值键之代碼。於步騾S77之判斷，判定被設定於變數 OKN之代碼非數值键之代碼（否之判定時），則移至步驟 S79，於此步騾S79做使游標移動之處理。另一方面，於步 騾S77之判斷，判定被設定於變數OKN之代碼為數值键之代 碼（是之判定時），此時確定已被輸入被顯示之表示候選文 字（顯示文字）作為輸入文字（步騾S78)。然後，移至步騾 S79，於此步騾S79做使游標移動之處理。步騾S79之處理結 4 Μ 83900 -17- 200405194 束，則移至步騾S76，將變數NKN之代碼設定於變數ΟΚΝ， 回到步騾S71。 於步騾S73之判斷，判定被設定於變數ΝΚΝ之代碼為數值 键之代碼（是之判定時），則再判斷被設定於變數ΝΚΝ之代碼 被設定於變數ΟΚΝ之代碼是否一致（步騾S80)。於此，判定 被設定於變數ΝΚΝ之代碼與被設定於變數ΟΚΝ之代碼不一 致（否之判定時），則移至步騾S81。於步騾S81，再判斷被設 定於變數ΟΚΝ之代碼是否為數值键之代碼。 於步驟S81之判斷，判定被設定於變數ΟΚΝ之代碼非數值 键之代碼（否之判定時），則於步驟S82使對應於被設定於變 數ΝΚΝ之代碼之輸入候選文字（第1候選）顯示於顯示裝置 lk，移至步騾S86。另一方面，於步騾S81之判斷，判定被設 定於變數OKN之代碼為數值鍵之代碼（是之判定時），於步騾 S83，將對應於現在作為輸入候選文字而被顯示之被設定於 變數OKN之代碼之文字確定作為輸入文字，以特定之樣態 使其顯示於顯示裝置lk。然後，於步騾S84，再使顯示於顯 示裝置lk之游標（於此，此游標係顯示輸入候選文字於被顯 示之位置者）顯示於下一個文字顯示位置，使對應於被設定 於變數NKN之代碼之輸入候選文字顯示於顯示裝置lk之對 應位置（游標位置）後，移至步騾S86。 另一方面，於步騾S80之判斷，判定被設定於變數NKN之 代碼與被設定於變數OKN之代碼一致（是之判定時），則此時 因同一键又被键碰觸，故將現在顯示之輸入候選文字變更 成下一個輸入候選文字（S85)。具體來說，例如現在顯示之 83900 -18- 200405194 輸入候選文字為「态」之情形，將此輸入候選文字變更成 「w」，再顯示。然後，移至步驟S86。 於以上之步騾S82、S84、S85之各階段顯示輸入候選文字’ 然而與此此輸入候選文字之顯示同時，於步驟S86，將對應 於該輸入候選文字之頻率參數及振幅參數與於特定之定時 將重設信號轉送至晋源裝置1 i内之FM音源24，使該輸入候 選文字之發音聲音合成、輸出。之後，於步驟S76將設定於 變數NKN之代碼設定於變數ONK後回到步騾S71，以後於文 字輸入模式之間，反覆以上之處理。以上說明了於文字輸 入模式之動作。 如此，於本實施型態可以使用同一音源裝置丨丨進行來電 等待模式時之來電旋律之播放及藉由文字輸入模式時之輸 入候選文字之聲音合成之播放。再者，於上述說明各動作 流程係一例，當然不限定於上述處理之流程。 於此，作為本貫施型態之實施例，將輸入候選文字之, 不例及其發首例顯示於圖8、9加以說明。圖8係假名 入時《一例。大字候選文字顯示於符號81所示之輸入候選 文2之輸入欄（假名漢字變換前之輸入欄）。同圖（a)顯示= 入可之狀態。此外，最終確定之文字顯示於符號82 : 、& ’使用者按「1」鍵，料輸人欄之游標（顯示 下線）位置顯示「志」之文字，並將其發音/a/聲音< :(圖啊。再者，使用者按Γι」鍵，則於輸入襴 、不位置顯示下—個文字之「、、」的文字，與此同 83900 -19- 200405194 則聲首合成、輸出（圖8(c))。其次，使用者按「6」键， 入、：幸二〈輸入候選文字<「、、」確定作為平假名之輸 一斿枯私動一個文字量。然後，於此位置顯示下一 個出入候選文字乏「 出（圖明。其次，使用」去、將其發音聲音合成、輸 — 使用者按「*」键，則於輸入欄之同一 2位置㈣、」作為輸人候選文字，將其發音/BAVA/ 口成‘出(圖8(e))。再者’按「*」鍵時之處理於圖7 ,、力乍'心私中，係於步驟S75之其他鍵處理被進行，此 形’為了進行聲音合成，步驟S75之處理後，非步驟S76 而是移到步驟S86。 傥明英又竽輸入時之一例（參照圖9)。於此例，輸 =候選=字顯示於符號91所示之游標位置。圖9⑷顯示輸入 、’、、狀心、首先，使用者按「2」鍵，則於游標位置顯示「A」 (文字並且作為其發晋將「之、、」，即爪―^ (發音符號 叫聲音合成、輸出。再者，使用者按％鍵，則於同一游 標位置顯示英文字「B」’作為其發音將「〜」，即/Β/ϋ (發音符號bi :)聲音合成、輸出。 以上參照圖面詳細說明了本發明之實施型態。當然，具 體結構不限於此實施型態，當然亦包含不脫離本發明要旨 ，範圍之結構等、。例如，不限於如上述之假名文字輸入或 英文字幸則入’北永話、上海話、廣東話、台灣話等中國話， 或韓語、德語、法語、西班牙語、葡萄牙語等其他國語亦 可以同樣貫犯。再者，於上述實施型態係將輸入候選文字 之發音聲音合成，惟利用電話功能輸入電話號碼之情形， 83900 -20- 200405194 碰觸鍵而輸入之輸入文字（此情形為號碼），非輸入候選文 字而是輸入文字本身，於此情形亦與輸入候選文字同樣地 將該輸入文字聲音合成其發音即可。 [發明之效果] 如以上所詳細說明，依據本發明，顯示與為輸入文字之 特定操作相對應之輸入候選文字時，聲音合成手段將該輸 入候選文字之發音聲音合成後輸出；藉此，使用該可攜式 終端裝置例如行動電話之使用者，因聽聲音合成之輸入候 選文字之發音可以確認該輸入候選文字，故無需如同先前 般地目視輸入候選文字之顯示來確認，便利性提高。此外， 即使使用者為視障者，文字輸入亦容易。再者，行動電話 之h形’將聲首合成手段使用於聲音合成之音源與使用於 可述行動電話所具備之來電音生成之音源共用，無需為了 聲骨合成手段而追加新裝置，可以抑制製造成本之增加。 【圖式簡單說明】 圖1為顯示本發明一實施型態之行動電話結構之圖。 圖2為顯示同實施型態之音源裝置結構成之方塊圖。 圖3為顯示包含於同實施型態之FM音源裝置之運算器結 構之方塊圖。 θ 圖4(a)圖4(b)為顯示於FM音源之運算器組合例之圖。 圖5為顯示藉由CSM聲音合成執行來電旋律合成之FMa 源結構之圖。 曰 圖6為來％等待模式時之動作流程圖。 圖7為文字輸入模式時之動作流程圖。 49〇 83900 -21 - 200405194 圖8(a)〜圖8(e)為顯示假名文字輸入時之輸入候選文字之 顯示例及其發音例之圖。 圖9(a)〜圖9(c)為顯示英文字輸入時之輸入候選文字之顯 示例及其發音例之圖。 圖10為顯示一般之行動電話之键（按鈕）之一例之圖。 圖11為顯示本發明一實施型態之行動電話外觀形狀之 圖。 【圖式代表符號說明】 1 行動電話（可攜式終端裝置） la CPU (聲音合成手段之一部分） lb ROM (聲立合成手段之一部分） lc RAM Id通信裝置 I e 輸入裝置 If 通話裝置 lg耳揚聲器 lh 麥克風 Π 音源裝置（聲音合成手段之一部分） lj 背面揚聲器 lk顯示裝置 II 天線 10 匯流排Set the pre-selected settings as ... 10,000, eight S62CPUL to the sound source device U. The music sequence data loaded with the music source is transferred to the caller melody, including 1 million yuan; after receiving the music sequence data, the synthesizer continues to play the caller melody. Next, when the anger button is judged to be on or the call button is judged to be off (be judged by the judge (when judged as YES)), return to = step: when Mei Road is disconnected, return to step S63. When it is not) 83900 -15- 493 200405194 On the other hand, when it is determined in step S63 that the call key is on (when judged to be YES), the CPU la gives the sound source device li a command to terminate the caller melody playback (+ step S65). At this stage, the source device Π terminates the playback of the incoming call melody. Then, it performs the processing during a normal call (step s66). In the next step S67, it is judged whether the end call key is on or off, and this judgment is repeated. Until the end call key is turned on (determined as 1). Then, when it is determined that the end call key is on (when determined = YES), the process proceeds to step S68. Then, the end of the call is performed in step S68. The processing at that time (line disconnection) returns to step S61. The above describes the operation from the incoming call to the line disconnection in the call waiting mode. Secondly, for the portable terminal device i of this embodiment configured in this way, The operation when inputting (character input mode) will be described with reference to the operation flowchart shown in the figure. Here, the portable terminal device 1 is in the character input mode through user-specific operations. In addition, the following and Figure 7 shows "NKN" (new key number) and "OKN" (old key number) as variables. "—" Is the cursor transport key. In addition, the code of the numeric key other than ^ is set in OKN as the initial value. In addition, the text input mode is used when e-mail, schedule, or other documents are created, or when an url is input during Internet connection, and the text input mode becomes necessary. First, in the step S71 determines whether the key has been touched (key on). Then, the judgment is repeated until the user touches the key. Here, the input device le detects the user's key touch, and when the key touch is detected, it indicates that the key has been touched. The touched key number is notified to the CPU la. From the input device k to the receipt of the key number notification, the CPU la judges that there is no key touch. Here, the key touch is detected (at step S71, it is 83900 -16- 200405194 of (Judgment). At this time, the CPU 1a receives the key number notification from the input device le, and sets the received key number to the variable NKN (step S72). Next, it determines whether the code (here, the key number) set to the variable NKN is Is the code of the numeric key (step S73). Here, if the code set to the variable NKN is not the code of the numeric key, move to step S74. At step S74, determine whether the code set to the variable NKN is Is the code of the cursor transport key ("4"). At the judgment of step S74, it is determined that the code set in the variable NKN is not the code of the cursor transport key (when the judgment is NO), then the execution of the corresponding other key corresponding to the other key is executed. Process (step S75). Then, it moves to step S76. In step S76, set the code of the variable NKN to the variable OKN, and then return to step S71. In addition, when the user's key operation is a specific mode change operation, that is, when the code set in the variable NKN is the code corresponding to the key of this mode change operation, the process of leaving the text input mode in FIG. 7 is terminated at step S75, and the process is terminated. Text input mode. On the other hand, in the judgment of step S74, it is judged that the code set in the variable NKN is the code of the cursor transport key (when the judgment is YES), and it moves to the next step S77. Then, in step S77, it is judged whether the code set in the variable OKN is the code of the numeric key. In the judgment of step S77, it is determined that the code of the variable OKN is set to a code other than a numeric key (when the judgment is no), the process moves to step S79, and at this step S79, the process of moving the cursor is performed. On the other hand, in the judgment of step S77, it is determined that the code set in the variable OKN is the code of the numeric key (when it is determined), and at this time, it is determined that the displayed candidate text (display text) is input as the input text. (Step S78). Then, it moves to step S79, and in this step S79, the process of moving the cursor is performed. If the processing of step S79 ends at 4 MU 83900 -17- 200405194, move to step S76, set the code of the variable NKN to the variable 0KN, and return to step S71. In the judgment of step S73, it is judged that the code set in the variable ΝΚΝ is the code of the numeric key (when the judgment is YES), and then it is judged whether the code set in the variable ΝΚΝ is set to the code of the variable 0KN (step S80 ). Here, it is determined that the code set in the variable NKN does not agree with the code set in the variable ΝΟΝ (when a negative determination is made), then the process proceeds to step S81. In step S81, it is judged whether the code set at the variable ΟΚΝ is the code of the numeric key. In the determination of step S81, it is determined that the code set in the variable 0KN is not a code of a numeric key (when the determination is no), and in step S82, the input candidate text (the first candidate) corresponding to the code set in the variable NGK is displayed. On the display device lk, go to step S86. On the other hand, in step S81, it is determined that the code set in the variable OKN is the code of the numeric key (when the determination is YES), and in step S83, the setting corresponding to the currently displayed input candidate text is set. The text of the code of the variable OKN is determined as the input text, and it is displayed on the display device lk in a specific state. Then, in step S84, the cursor displayed on the display device lk (here, this cursor is a display of the input candidate text at the position to be displayed) is displayed at the next text display position, corresponding to the variable NKN set. After the input candidate text of the code is displayed at the corresponding position (cursor position) of the display device lk, move to step S86. On the other hand, in the judgment of step S80, it is determined that the code set in the variable NKN is consistent with the code set in the variable OKN (when it is determined), then the same key is touched again by the key, so it will be The displayed input candidate text is changed to the next input candidate text (S85). Specifically, for example, the 83900 -18- 200405194 input candidate text currently displayed is "state", and the input candidate text is changed to "w" and then displayed. Then, the process proceeds to step S86. In the above steps, S82, S84, and S85 are displayed as input candidate characters. However, at the same time as the input candidate characters are displayed, in step S86, the frequency parameters and amplitude parameters corresponding to the input candidate characters are compared with the specific ones. The reset signal is regularly transmitted to the FM sound source 24 in the Jin source device 1 i to synthesize and output the pronunciation sound of the input candidate text. After that, in step S76, the code set in the variable NKN is set to the variable ONK, and then returns to step S71. After that, the above processing is repeated between the text input modes. The above explains the operation in the text input mode. In this way, in the embodiment, the same sound source device can be used to play the caller melody in the call waiting mode and the sound synthesis by inputting the candidate text in the text input mode. It should be noted that each operation flow described above is an example, and is not limited to the above-mentioned processing flow. Here, as an embodiment of the present embodiment, examples of input candidate characters, and examples thereof are shown in FIGS. 8 and 9 for explanation. Figure 8 is a case of "Kana" when entering. Large-character candidate text is displayed in the input field of input candidate text 2 (input field before Kana Kanji conversion) shown by symbol 81. Shown in the same figure (a) = the state of being available. In addition, the finalized text is displayed on the symbol 82 :, & 'The user presses the "1" key, and the text of "Chi" is displayed at the cursor (displayed off-line) position in the input field, and it is pronounced / a / sound <: (Picture. Moreover, if the user presses the Γι ”key, the text of the next character," ,, "is displayed without inputting 襕, as with 83900 -19- 200405194. (Figure 8 (c)). Secondly, the user presses the "6" key, and enters: Xingji <Enter the candidate text < ",," to determine the amount of input as a hiragana input. This position shows that the next entry candidate text lacks "out" (pictured. Next, use "" to synthesize its pronunciation sound and input — the user presses the "*" key, then the same two positions in the input field: ㈣, "as input. Candidate characters are pronounced / BAVA / into 'out' (Figure 8 (e)). Furthermore, the processing when '*' key is pressed is shown in Figure 7, and Licha's heart is in step S75. Other key processing is performed. This shape is used to synthesize sound. After the processing in step S75, it is not step S76 but moves to step S86. An example when Mingying is inputting again (refer to Figure 9). In this example, the input = candidate = character is displayed at the cursor position shown by symbol 91. Figure 9 shows the input, ',, shape, first, use If you press the "2" key, "A" (the text will be displayed at the cursor position, and "zhi", "", which is a claw ^ (pronounced symbol is called sound synthesis and output. Furthermore, the user presses the% key , The English word "B" 'is displayed at the same cursor position as its pronunciation, and "~", that is, / Β / ϋ (pronounced symbol bi :) is synthesized and outputted. The above describes the implementation mode of the present invention in detail with reference to the drawings. . Of course, the specific structure is not limited to this implementation type, and of course includes the structure without departing from the gist of the present invention. For example, it is not limited to the input of kana characters or the English characters such as' North Yonghua, Shanghai Dialect, Cantonese, Taiwanese and other Chinese languages, or Korean, German, French, Spanish, Portuguese and other national languages can be equally committed. In addition, in the above implementation type, the pronunciation of the input candidate text is synthesized, but using phone In the case where a phone number can be entered, 83900 -20- 200405194 The input text entered by touching the key (the number in this case) is not the candidate text but the text itself. In this case, the input is the same as the input candidate text. The text sound can be synthesized by its pronunciation. [Effects of the Invention] As explained in detail above, according to the present invention, when the input candidate text corresponding to a specific operation for inputting text is displayed, the sound synthesis means generates the pronunciation sound of the input candidate text. Output after synthesizing; thereby, the user using the portable terminal device, such as a mobile phone, can confirm the input candidate text by listening to the pronunciation of the input candidate text synthesized by the voice, so there is no need to visually display the input candidate text as before. To confirm, convenience is improved. In addition, even if the user is visually impaired, text input is easy. In addition, the h-shape of the mobile phone uses the sound synthesis method for the sound synthesis and the sound source used for the call tone generation provided by the described mobile phone. There is no need to add a new device for the sound bone synthesis method, which can suppress Increase in manufacturing costs. [Brief Description of the Drawings] FIG. 1 is a diagram showing the structure of a mobile phone according to an embodiment of the present invention. FIG. 2 is a block diagram showing the structure of a sound source device in the same embodiment. Fig. 3 is a block diagram showing the structure of an arithmetic unit included in the FM sound source device of the same embodiment. θ Fig. 4 (a) and Fig. 4 (b) are diagrams showing an example of a combination of calculators displayed on an FM sound source. FIG. 5 is a diagram showing the structure of a FMa source that performs call melodic synthesis by CSM sound synthesis. Fig. 6 is an operation flowchart in the% waiting mode. FIG. 7 is an operation flowchart in the text input mode. 49〇 83900 -21-200405194 Figures 8 (a) to 8 (e) are diagrams showing examples of display of input candidate characters and pronunciation examples when inputting kana characters. Figs. 9 (a) to 9 (c) are diagrams showing examples of display of input candidate characters and pronunciation examples when inputting English characters. FIG. 10 is a diagram showing an example of keys (buttons) of a general mobile phone. Fig. 11 is a diagram showing the appearance of a mobile phone according to an embodiment of the present invention. [Illustration of Representative Symbols] 1 Mobile phone (portable terminal device) la CPU (part of sound synthesis means) lb ROM (part of sound synthesis means) lc RAM Id communication device I e Input device If call device lg ear Speaker lh microphone Π sound source device (part of sound synthesis means) lj rear speaker lk display device II antenna 10 bus
21 輸出入I/F21 I / F
22 FIFO 83900 -22- 200405194 23排序器 24 FM音源 25 WT音源 26 波形記憶體 27 加法器 30 運算器 31 SIN波形表 32 相位產生器（PG) 33 加法器 34 色絡產生器（EG) 35 乘法器 50 加法器 -23 - 遲 8390022 FIFO 83900 -22- 200405194 23 Sequencer 24 FM source 25 WT source 26 Wave memory 27 Adder 30 Arithmetic unit 31 SIN waveform table 32 Phase generator (PG) 33 Adder 34 Color network generator (EG) 35 Multiplication 50 adder-23-late 83900