Lafal buatan (speech synthesis) adalah penghasilan pertuturan manusia tanpa mengggunakan suara manusia secara langsung.
Secara umum, pelafal buatan (speech synthesizer) adalah perisian atau perkakasan yang mampu menghasilkan ujaran buatan (artificial speech). Sistem ujaran buatan "Speech synthesis" sering dipanggil sistem teks-ke-pertuturan (text-to-speech TTS) merujuk kepada keupayaannya untuk menukar teks kepada pertuturan. Bagaimanapun, terdapat sistem yang hanya menghasilkan wakil simbol linguistik "symbolic linguistic representation" seperti transkripsi fonetis "phonetic transcription" kepada pertuturan.
Seimbas berkenaan teknologi Lafal Buatan (Speech Synthesis) [sunting]
Sistem teks-ke-pertuturan (atau engin) terdiri daripada dua bahagian: bahagian depan dan bahagian belakang. Umumnya, bahagian hadapan mengambil input dalam bentuk teks dan output wakil simbol linguistik "symbolic linguistic representation". Bahagian belakang mengambil wakil simbol linguistik "symbolic linguistic representation" sebagai input dan menghasilkan lafal buatan waveform. naturalness pensintesis pertuturan biasanya merujuk kepada berapa tepat bunyi output kedengaran seperti manusia sebenar.
Bahagian hadapan mempunyai dua tugas utama. Pertama ia mengambil teks mentah dan menukar sebahagian daripadanya seperti nombor dan ringkasan kepada perkataan bertulis yang setara. Proses ini dikenali sebagai menormalkan teks (text normalization), pre-processing, atau tokenization. Kemudian ia memberikan transkripsi fonetis "phonetic transcription" kepada setiap perkataan, dan menandakan teks kepada pelbagai unit prosodi, seperti farsa, klaus "clauses", dan ayat. Proses memberikan transkripsi fonetis kepada perkataan ini dikenali sebagai teks-ke-fonem (text-to-phoneme TTP) atau penukaran grafem-ke-fonem (grapheme-to-phoneme GTP). Gabungan transkripsi fonetis dan maklumat mengenai unit prosodi membentuk output wakil simbol linguistik pada bahagian hadapan.
Bahagian lain, bahagian belakang, mengambil wakil simbol linguistik dan menukarkannya kepada output bunyi sebenar. Bahagian belakang sering dirujuk sebagai pensintesis. Teknik pensintesis yang berlainan dibincangkan di bawah.
Sejak awal lagi sebelum pemproses signal eletronik moden dicipta, penyelidik pertuturan cuba membina mesin yang menghasilkan pertuturan manusia. Contoh awal 'kepala bercakap' dibuat oleh Gerbert of Aurillac (m. 1003), Albertus Magnus (1198-1280), dan Roger Bacon (1214-1294).
Pada tahun 1779, Christian Kratzenstein dari St. Petersburg membina model peti suara manusia yang mampu menghasilkan lima bunyi vowel panjang (a, e, i, o dan u). Ini diikuti dengan 'Mesin Pertuturan Mekanikal Akustik - Acoustic-Mechanical Speech Machine' berkuasa penghembus "bellows-operated" oleh Wolfgang von Kempelen dari Vienna, Austria, yang digambarkan dalam kertas kerjanya pada tahun 1791 Mechanismus der menschlichen Sprache nebst der Beschreibung seiner sprechenden Maschine (J.B. Degen, Wien). Mesin ini menambahkan model lidah dan bibir, membolehkan ia menghasilkan bunyi consonant dan vowels. Pada tahun 1837 Charles Wheatstone menghasilkan 'mesin bertutur' berasaskan reka bentuk von Kempelen, dan pada tahun 1857 M. Faber membina 'Euphonia'. Reka bentuk Wheatstone dihidupkan kembali pada tahun 1923 by Paget.
|Bell Labs VODER|
Pada tahun 1930s, Bell Labs memajukan VOCODER, penganalisa dan pensyintesis eletronik menggunakan papan kekunci yang dikatakan jelas difahami. Homer Dudley memajukan lagi peranti ini kepada VODER, yang dipamernya di pesta Dunia New York 1939 (1939 New York World's Fair).
Pensintesis pertuturan awal berbunyi seperti robot dan sering sukar difahami. Output dari sistem TTS terkini kadang-kala sukar dibezakan dengan pertuturan manusia sebenar.
Sungguhpun dengan kejayaan sintesis pertuturan eletronik, penyelidikan masih dilakukan mengenai sintesis pertuturan eletronik bagi kegunaan robot seperti manusia. Sungguhpun sintesis pertuturan eletronik sempurna dihadkan oleh kualiti transducer (biasanya pembesar suara) yang menghasilkan bunyi, dengan itu system mekanikal robot mungkin mampu menghasilkan bunyi lebih normal berbanding pembesar suara kecil.
Pensintesis pertuturan eletronik berasaskan computer pertama dicipta pada akhir 1950-an dan system teks kepada pertuturan lengkap pertama disiapkan pada 1968. Sejak itu, terdapat banyak kemajuan dalam teknologi yang digunakan bagi penghasilan pertuturan eletronik, dan system teks-kepada-pertuturan moden sering kali mampu menghasilkan bunyi yang sukar dibezakan berbanding pertuturan manusia sebenar. Lihat #Contoh system kini di bawah bagi system teks-kepada-pertuturan perdagangan tercanggih dan yang bebas guna.
- Dennis Klatt's History of Speech Synthesis
- History and Development of Speech Synthesis, from Helsinki University of Technology Laboratory of Acoustics and Audio Signal Processing (2003-09-13)
Teknologi-teknologi Buatan [sunting]
Terdapat dua tekologi utama yang digunaka bagi menghasilkan bentuk gelombang pertuturan buatan: sintetik jeraitan dan sintesis formant
Sintetik jeraitan ("Concatenative synthesis ") [sunting]
Sintetik jeraitan adalah berasaskan penyatuan (atau jeraitan) bahagian-bahagian pertuturan yang dirakam. Biasanya, sintetik jeraitan memberikan pertuturan sintesis paling hidup. Tetapi variasi normal dalam pertuturan dan teknik automatik bagi memecah bentuk geombang kadang-kala menghasilkan herotan boleh dikesan dalam output, mengganggu bunyi seperti hidup. Terdapat tiga jenis kecil sintetik jeraitan.
- Sintesis unit pilihan menggunakan sejumlah besar pangkalan data (lebih daripada sejam pertuturan yang di rakam). emasa ciptaan pangkalan data, setiap sebutan dirakam adalah bahagian yang membentuk sebahagian atau keseluruhan berikut: individu talifon, syllables, morphemes, words, phrases, and sentences. The division into segments can be done using a number of techniques, like clustering, using a specially modified speech recognizer, or by hand, using visual representations such as the waveform and spectrogram. An index of the units in the speech database is then created based on the segmentation and acoustic parameters like the fundamental frequency (pitch). At runtime, the desired target utterance is created by determining the best chain of candidate units from the database (unit selection). This technique gives the greatest naturalness due to the fact that it does not apply digital signal processing techniques to the recorded speech, which often makes recorded speech sound less natural. In fact, output the best unit selection systems are often indistinguishable from real human voices, especially in contexts for which the TTS system has been tuned. However, maximum naturalness often requires unit selection speech databases to be very large, in some systems ranging into the gigabytes of recorded data and numbering into the dozens of hours of recorded speech.
- Diphone synthesis uses a minimal speech database containing all the Diphones (sound-to-sound transitions) occurring in a given language. The number of diphones depends on the phonotactics of the language: Spanish has about 800 diphones, German about 2500. In diphone synthesis, only one example of each diphone is contained in the speech database. At runtime, the target prosody of a sentence is superimposed on these minimal units by means of digital signal processing techniques such as Linear predictive coding, PSOLA or MBROLA. The quality of the resulting speech is generally not as good as that from unit selection but more natural-sounding than the output of formant synthesizers. Diphone synthesis suffers from the sonic glitches of concatenative synthesis and the robotic-sounding nature of formant synthesis, and has few of the advantages of either approach other than small size. As such, its use in commercial applications is declining, although it continues to be used in research because there are a number of freely available implementations.
- Domain-specific synthesis concatenates pre-recorded words and phrases to create complete utterances. It is used in applications where the variety of texts the system will output is limited to a particular domain, like transit schedule announcements or weather reports. This technology is very simple to implement, and has been in commercial use for a long time: this is the technology used by things like talking clocks and calculators. The naturalness of these systems can potentially be very high because the variety of sentence types is limited and closely matches the prosody and intonation of the original recordings. However, because these systems are limited by the words and phrases in its database, they are not general-purpose and can only synthesize the combinations of words and phrases they have been pre-programmed with.
Formant synthesis [sunting]
Formant synthesis does not use any human speech samples at runtime. Instead, the output synthesized speech is created using an acoustic model. Parameters such as fundamental frequency, voicing, and noise levels are varied over time to create a waveform of artificial speech. This method is sometimes called Rule-based synthesis but some argue that because many concatenative systems use rule-based components for some parts of the system, like the front end, the term is not specific enough.
Many systems based on formant synthesis technology generate artificial, robotic-sounding speech, and the output would never be mistaken for the speech of a real human. However, maximum naturalness is not always the goal of a speech synthesis system, and formant synthesis systems have some advantages over concatenative systems.
Formant synthesized speech can be very reliably intelligible, even at very high speeds, avoiding the acoustic glitches that can often plague concatenative systems. High speed synthesized speech is often used by the visually impaired for quickly navigating computers using a screen reader. Second, formant synthesizers are often smaller programs than concatenative systems because they do not have a database of speech samples. They can thus be used in embedded computing situations where memory space and processor power are often scarce. Last, because formant-based systems have total control over all aspects of the output speech, a wide variety of prosody or intonation can be output, conveying not just questions and statements, but a variety of emotions and tones of voice.
Other synthesis methods [sunting]
- Articulatory synthesis is a synthesis method mostly of academic interest at the moment. It is based on computational models of the human vocal tract and the articulation processes occurring there. These models are currently not sufficiently advanced or computationally efficient to be used in commercial speech synthesis systems.
- Hybrid synthesis marries aspects of formant and concatenative synthesis to minimize the acoustic glitches when speech segments are concatenated.
Text normalization challenges [sunting]
The process of normalizing text is rarely straightforward. Texts are full of homographs, numbers and abbreviations that all ultimately require expansion into a phonetic representation.
There are many words in English which are pronounced differently based on context. Some examples:
- project: My latest project is to learn how to better project my voice.
- bow: The girl with the bow in her hair was told to bow deeply when greeting her superiors.
Most TTS systems do not generate semantic representations of their input texts, as processes for doing so are not reliable, well-understood, or computationally effective. As a result, various heuristic techniques are used to guess the proper way to disambiguate homographs, like looking at neighboring words and using statistics about frequency of occurrence.
Deciding how to convert numbers is another problem TTS systems have to address. It is a fairly simple programming challenge to convert a number into words, like 1325 becoming "one thousand three hundred twenty-five". However, numbers occur in many different contexts in texts, and 1325 should probably be read as "thirteen twenty-five" when part of an address (1325 Main St.) and as "one three two five" if it is the last four digits of a social security number. Often a TTS system can infer how to expand a number based on surrounding words, numbers, and punctuation, and sometimes the systems provide a way to specify the type of context if it is ambiguous.
Similarly, abbreviations like "etc." are easily rendered as "et cetera", but often abbreviations can be ambiguous. For example, the abbreviation "in." in the following example: "Yesterday it rained 3 in. Take 1 out, then put 3 in." "St." can also be ambiguous: "St. John St." TTS systems with intelligent front ends can make educated guesses about how to deal with ambiguous abbreviations, while others do the same thing in all cases, resulting in nonsensical but sometimes comical outputs: "Yesterday it rained three in." or "Take one out, then put three inches."
Text-to-phoneme challenges [sunting]
Speech synthesis systems use two basic approaches to determine the pronunciation of a word based on its spelling, a process which is often called text-to-phoneme or grapheme-to-phoneme conversion, as phoneme is the term used by linguists to describe distinctive sounds in a language.
The simplest approach to text-to-phoneme conversion is the dictionary-based approach. In this approach, a large dictionary containing all the words of a language and their correct pronunciation is stored by the program. Determining the correct pronunciation of each word is a matter of looking up each word in the dictionary and replacing the spelling with the pronunciation specified in the dictionary.
The other approach used for text-to-phoneme conversion is the rule-based approach. In this approach, rules for the pronunciations of words are applied to words to work out their pronunciations based on their spellings. This is similar to the "sounding out" approach to learning reading.
Each approach has advantages and drawbacks. The dictionary-based approach has the advantages of being quick and accurate, but it completely fails if it is given a word which is not in its dictionary, and as dictionary size grows, so too does the memory space requirements of the synthesis system. On the other hand, the rule-based approach works on any input, but the complexity of the rules grows substantially as it takes into account irregular spellings or pronunciations. As a result, nearly all speech synthesis systems use a combination of both approaches.
Some languages, like Spanish, have a very regular writing system, and the prediction of the pronunciation of words based on the spelling works correctly in nearly all instances. Speech synthesis systems for languages like this often use the rule-based approach as the core approach for text-to-phoneme conversion, resorting to dictionaries only for those few words, like foreign names and borrowings, whose pronunciation is not obvious from the spelling. On the other hand, speech synthesis for languages like English, which have extremely irregular spelling systems, often rely mostly on dictionaries and use rule-based approaches only for unusual words or names that aren't in the dictionary.
Examples of current systems [sunting]
Some freely available text-to-speech systems:
- Festival is a freely available complete diphone concatenation and unit selection TTS system.
- Flite (Festival-lite) is a smaller, faster alterative version of Festival designed for embedded systems and high volume servers.
- MBROLA is a freely available diphone concatenation system (back end).
- Gnuspeech is an extensible, text-to-speech package, based on real-time, articulatory, speech-synthesis-by-rules.
Some very natural sounding commercial concatenative TTS systems with online demos: All of these have US English, most have other languages available.
- Speechworks Speechify
- Rhetorical rVoice
- Loquendo TTS
- ScanSoft RealSpeak
- Nuance Vocalizer
- AT&T Natural Voices
Well known external hardware devices:
- Double Talk PC
Recently Available Hardware devices:
Speech synthesis markup languages [sunting]
A number of markup languages for rendition of text as speech in an XML compliant format, have been established, most recently the SSML proposed by the W3C (still in draft status at the time of this writing). Older speech synthesis markup languages include SABLE and JSML. Although each of these was proposed as a new standard, still none of them has been widely adopted.
Speech synthesis markup languages should be distinguished from dialogue markup languages such as VoiceXML, which includes, in addition to text-to-speech markup, tags related to speech recognition, dialogue management and touchtone dialing.
Pautan luar [sunting]
- Contoh system perdagangan TTS.
- Free sistem Lafal Buatan bagi mereka yang mengalami permasaalahan pertuturan, dengan pautan bagi teknologi berkait pertuturan lain dan sumber bagi PALS.
- Text to Speech dalam Bahasa