Download presentation
Presentation is loading. Please wait.
1
Automatic Speech Recognition
Lecture 7 Automatic Speech Recognition CS 4705
2
What is speech recognition?
Transcribing words? Understanding meaning? Today: Overview ASR issues Building an ASR system Using an ASR system Future research
3
“It’s hard to ... recognize speech/wreck a nice beach”
Speaker variability: within and across Recording environment varies wrt noise Transcription task must handle all of this and produce a transcript of what was said, from limited, noisy information in the speech signal Success: low word error rate (WER) WER = (S+I+D)/N * 100 Thesis test vs. This is a test. 75% WER Understanding task must do more: from words to meaning
4
Measure concept accuracy (CA) of string in terms of accuracy of recognition of domain concepts mentioned in string and their values I want to go from Boston to Baltimore on September 29 Domain concepts Values source city Boston target city Baltimore travel date September 29 Score recognized string “Go from Boston to Washington on December 29” (1/3 = 33% CA) “Go to Boston from Baltimore on September 29”
5
Again, the Noisy Channel Model
Source Noisy Channel Decoder Input to channel: spoken sentence s Output from channel: an observation O Decoding task: find s = P(s|O) Using Bayes Rule And since P(O) doesn’t change for any hypothetical s’ s’ = P(O|s) P(s) P(O|s) is the observation likelihood, or Acoustic Model, and P(s) is the prior, or Language Model
6
What do we need to build use an ASR system?
Corpora for training and testing of components Feature extraction component Pronunciation Model Acoustic Model Language Model Algorithms to search hypothesis space efficiently
7
Training and Test Corpora
Collect corpora appropriate for recognition task at hand Small speech + phonetic transcription to associate sounds with symbols (Acoustic Model) Large (>= 60 hrs) speech + orthographic transcription to associate words with sounds (Acoustic Model) Very large text corpus to identify unigram and bigram probabilities (Language Model)
8
Representing the Signal
What parameters (features) of the speech input Can be extracted automatically Will preserve phonetic identity and distinguish it from other phones Will be independent of speaker variability and channel conditions Will not take up too much space Speech representations (for [ae] in had): Waveform: change in sound pressure over time LPC Spectrum: component frequencies of a waveform Spectrogram: overall view of how frequencies change from phone to phone
9
Signal divided into frames
Speech captured by microphone and sampled (digitized) -- may not capture all vital information Signal divided into frames Power spectrum computed to represent energy in different bands of the signal LPC spectrum, Cepstra, PLP Each frame’s spectral features represented by small set of numbers Frames clustered into ‘phone-like’ groups (phones in context) -- Gaussian or other models
10
Why this works? Different phonemes have different spectral characteristics Why it doesn’t work? Phonemes can have different properties in different acoustic contexts, spoken by different people … Nice white rice
11
Pronunciation Model Models likelihood of word given network of candidate phone hypotheses (weighted phone lattice) Allophones: butter vs. but Multiple pronunciations for each word Lexicon may be weighted automaton or simple dictionary Words come from all corpora; pronunciations from pronouncing dictionary or TTS system
12
Acoustic Models Model likelihood of phones or subphones given spectral features and prior context Use pronunciation models Usually represented as HMM Set of states representing phones or other subword units Transition probabilities on states: how likely is it to see one phone after seeing another? Observation/output likelihoods: how likely is spectral feature vector to be observed from phone state i, given phone state i-1?
13
Initial estimates for Transition probabilities between phone states Observation probabilities associating phone states with acoustic examples Re-estimate both probabilities by feeding the HMM the transcribed speech training corpus (forced alignment) I.e., we tell the HMM the ‘right’ answers -- which words to associate with which sequences of sounds Iteratively retrain the transition and observation probabilities by running the training data through the model and scoring output until no improvement
14
Language Model Models likelihood of word given prior word and of entire sentence Ngram models: Build the LM by calculating bigram or trigram probabilities from text training corpus Smoothing issues very important for real systems Grammars Finite state grammar or Context Free Grammar (CFG) or semantic grammar Out of Vocabulary (OOV) problem
15
Entropy H(X): the amount of information in a LM, grammar
How many bits will it take on average to encode a choice or a piece of information? More likely things will take fewer bits to encode Perplexity 2H: a measure of the weighted mean number of choice points in e.g. a language model
16
Search/Decoding Find the best hypothesis P(O|s) P(s) given
Lattice of subword units (Acoustic Model) Segmentation of all paths into possible words (Pronunciation Model) Probabilities of word sequences (Language Model) Produces a huge search space: How to reduce? Lattice minimization and determinization Forward algorithm: sum of all paths leading to a state Viterbi algorithm: max of all paths leading to a state
17
Forward-backward (Baum-Welch, Expectation-Maximization) algorithm: computes probability of sequence at any state in search space Beam search: prune the lattice
18
Varieties of Speech Recognition
Telephone, microphones Input device > 30dB (high) < 10dB (low) SNR <10 > 100 Perplexity Finite state, ngrams, CFGs, CSGs Language Model <20 5K --> 60K -->~1M Vocabulary size Speaker-dependent or independent Enrollment Read, prepared, spontaneous Style Isolated words continuous Mode
19
Challenges for Transcription
Robustness to channel characteristics and noise Portability to new applications Adapatation: to speakers, to environments LMs: simple ngrams need help Confidence measures OOV words New speaking styles/genres New applications
20
Challenges for Understanding
Recognizing communicative ‘problems’ ASR errors User corrections Disfluencies and self-repairs Possibilities: Recognizing speaker emotion Identifying speech acts: okay Locating topic boundaries for topic tracking, audio browsing, speech data mining
21
An Unsuccessful Dialogue
S: Hi, this is AT&T Amtrak schedule system. This is TOOT. How may I help you? U: I'd like to find a train from Ch- to Chicago from Baltimore. S: Sorry, I can't understand you. Please repeat your utterance. U: I'd like to find a train from Baltimore to Chicago. S: I heard you say “leave today around 5 thirty in the morning .“ Which city do you want to go to? U: I would like to leave Saturday at eight in the morning. Note that the previous speaker was not a native speaker of English -- non-native speakers are often harder for recognizers to understand. Here however is a native speaker who has a terrible time with TOOT -- which never seems to understand ‘Baltimore’. Note the different ways she produces that city name during the dialogue. You’ll here some clear examples of hyperarticulation. $TOOT/non_adapt/MixedImplicit/award/ -- TOOT can never understand Baltimore
22
S: You can say the arrival city name, such as “New York City."
U: Baltimore to Chicago. S: You can say the arrival city name, such as “New York City.” U: Baltimore to Chicago. (I hate it.) U: Train from Baltimore ...S: ...I heard you s...U: to Chicago.... Hello? S: You can say the name of your departure city, such as "New York City." U: Baltimore. Note how she changes her pronunciation of “Baltimore”, hoping TOOT will finally here it clearly enough!
23
U: Bal-ti-more.... S: You can say... U: Baltimore. S: Do you want to continue and find a new set of trains? U: Yes. S: You can say "yes" or "I am done here" to find a new set of trains. S: Thank you for using AT&T Amtrak train time table system. See you next time. U: I nev-
24
Summary ASR technology relies upon a large number of phenomena and techniques we’ve already seen to convert sound into words Phonetic/phonological, morphological, and lexical events FSA’s, Ngrams, Dynamic programming algorithms Better modeling of linguistic phenomena will be needed to improve performance on transcription and especially on understanding For next class: we’ll start talking about larger structures in language above the word (Ch 8)
25
Disfluencies and Self-Repairs
Disfluencies abound in spontaneous speech every 4.6s in radio call-in (Blackmer & Mitton ‘91) hesitation: Ch- change strategy. filled pause: Um Baltimore. self-repair: Ba- uh Chicago. Hard to recognize Ch- change strategy. --> to D C D C today ten fifteen. Um Baltimore. --> From Baltimore ten. Ba- uh Chicago. --> For Boston Chicago. Kasl & Mahl: 41% more filled pauses in audio only vs ftf; Oviatt: 8.83 to 5.50% disfluencies in phone conversations vs. non /n/u118/exp98/adapt/MixedImplicit/mccoy/task1 line 198, rehesitation (mor- morning --> not really) /n/u118/exp98/adapt/UserNoConfirm/sgoel/task1 line 1315, filled pause (um baltimore --> from baltimoreten) /n/u118/exp98/adapt/UserNoConfirm/selina/task1, line 1132, repair (Ba- uh Chicago --> for Boston Chicago)
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.