This paper proposes a unified framework, All-in-One ASR, that allows a single model to support multiple automatic speech recognition (ASR) paradigms, including connectionist temporal classification (CTC), attention-based encoder-decoder (AED), and Transducer, in both offline and streaming modes. While each ASR architecture offers distinct advantages and trade-offs depending on the application, maintaining separate models for each scenario incurs substantial development and deployment costs. To address this issue, we introduce a multi-mode joiner that enables seamless integration of various ASR modes within a single unified model. Experiments show that All-in-One ASR significantly reduces the total model footprint while matching or even surpassing the recognition performance of individually optimized ASR models. Furthermore, joint decoding leverages the complementary strengths of different ASR modes, yielding additional improvements in recognition accuracy.

In recent advancements in end-to-end ASR, large-scale self-supervised or weakly supervised models have achieved a significant milestone. However, it remains challenging to train consistently high-performing multilingual models, transferable to languages without much resource. In this study, we propose embedding universal phonological knowledge to multilingual ASR by predicting international phonetic alphabet (IPA) targets and universal articulatory features alongside primary grapheme targets. These additions are expected to provide effective inductive bias or regularization for predicting grapheme targets across various languages. In the experiments, which involve fine-tuning a pre-trained XLS-R model using 10,400 hours of data across 120 languages from the Common Voice corpus, our proposed method achieved a 6.81% relative reduction in character error rate.

This paper explores enhancing streaming speech recognition through the integration of chunk-wise attention and selective state space models (SSMs). The proposed framework replaces the quadratic complexity of attention-based context incorporation with a fully recurrent module based on selective SSMs. Leveraging its constant-time inference and selective reasoning capability, our recurrent module efficiently takes advantage of the entire history across chunks, while conventional chunk-wise encoders are typically constrained to a fixed-length left context to ensure real-time performance. Meanwhile, our within-chunk attention module is designed to focus exclusively on local features within short speech segments, thereby complementing the autoregressive behavior of the SSM. Experimental results from RNN Transducer-based streaming speech recognition indicate that the recurrent and attention modules work synergistically, yielding lower error rates and reduced runtime costs compared to baseline Conformer systems.

Extending the RNN Transducer (RNNT) to recognize multi-talker speech is essential for wider automatic speech recognition (ASR) applications. Multi-talker RNNT (MT-RNNT) aims to achieve recognition without relying on costly front-end source separation. MT-RNNT is conventionally implemented using architectures with multiple encoders or decoders, or by serializing all speakers’ transcriptions into a single output stream. The first approach is computationally expensive, particularly due to the need for multiple encoder processing. In contrast, the second approach involves a complex label generation process, requiring accurate timestamps of all words spoken by all speakers in the mixture, obtained from an external ASR system. In this paper, we propose a novel alignment-free training scheme for the MT-RNNT (MT-RNNT-AFT) that adopts the standard RNNT architecture. The target labels are created by appending a prompt token corresponding to each speaker at the beginning of the transcription, reflecting the order of each speaker’s appearance in the mixtures. Thus, MT-RNNT-AFT can be trained without relying on accurate alignments, and it can recognize all speakers’ speech with just one round of encoder processing. Experiments show that MT-RNNT-AFT achieves performance comparable to that of the state-of-the-art alternatives, while greatly simplifying the training process.

While waveform-domain speech enhancement (SE) has been extensively investigated in recent years and achieves state-of-the-art performance in many datasets, spectrogram-based SE tends to show robust and stable enhancement behavior. In this paper, we propose a waveform-spectrogram hybrid method (WaveSpecEnc) to improve the robustness of waveform-domain SE. WaveSpecEnc refines the corresponding temporal feature map by spectrogram encoding in each encoder layer. Incorporating spectral information provides robust human hearing experience performance. However, it has a minor automatic speech recognition (ASR) improvement. Thus, we improve it for robust ASR by further utilizing spectrogram encoding information (WaveSpecEnc+) to both the SE front-end and ASR back-end. Experimental results using the CHiME-4 dataset show that ASR performance in real evaluation sets is consistently improved with the proposed method, which outperformed others, including DEMUCS and Conv-Tasnet. Refining in the shallow encoder layers is very effective, and the effect is confirmed even with a strong ASR baseline using WavLM.

Time-domain speech enhancement (SE) has recently been intensively investigated. Among recent works, DEMUCS [1] introduces multi-resolution STFT loss to enhance performance. However, some resolutions used for STFT contain non-stationary signals, and it is challenging to learn multi-resolution frequency losses simultaneously with only one output. For better use of multi-resolution frequency information, we supplement multiple spectrograms in different frame lengths into the time-domain encoders. They extract stationary frequency information in both narrowband and wideband. We also adopt multiple decoder outputs, each of which computes its corresponding resolution frequency loss. Experimental results show that (1) it is more effective to fuse stationary frequency features than non-stationary features in the encoder, and (2) the multiple outputs consistent with the frequency loss improve performance. Experiments on the Voice-Bank dataset show that the proposed method obtained a 0.14 PESQ improvement.

While speech emotion recognition (SER) has been actively studied, the amount and variations of training data are limited compared with speech recognition and speaker recognition tasks. Therefore, it is promising to combine multiple corpora to train a generalized SER model. However, the manner of emotion expression is different according to the settings, task domains, and languages. In particular, there is a mismatch between acted datasets and spontaneous datasets since the former includes much more rich and explicit emotion expressions than the latter. In this paper, we investigate effective combination methods based on multi-task learning (MTL) considering the style attribute. We also hypothesize the neutral expression, which has the largest number of samples, is not affected by the style, and thus propose a selective MTL method that applies MTL to emotion categories except for the neutral category. Experimental evaluations using the IEMOCAP database and a call center dataset confirm the effect of the combination of the two corpora, MTL, and the proposed selective MTL.

In automatic speech recognition (ASR) rescoring, the hypothesis with the fewest errors should be selected from the$n$-best list using a language model (LM). However, LMs are usually trained to maximize the likelihood of correct word sequences, not to detect ASR errors. We propose an ASR rescoring method for directly detecting errors with ELECTRA, which is originally a pre-training method for NLP tasks. ELECTRA is pre-trained to predict whether each word is replaced by BERT or not, which can simulate ASR error detection on large text corpora. To make this pre-training closer to ASR error detection, we further propose an extended version of ELECTRA called phone-attentive ELECTRA (P-ELECTRA). In the pre-training of P-ELECTRA, each word is replaced by a phone-to-word conversion model, which leverages phone information to generate acoustically similar words. Since our rescoring method is optimized for detecting errors, it can also be used for word-level confidence estimation. Experimental evaluations on the Librispeech and TED-LIUM2 corpora show that our rescoring method with ELECTRA is competitive with conventional rescoring methods with faster inference. ELECTRA also performs better in confidence estimation than BERT because it can learn to detect inappropriate words not only in fine-tuning but also in pre-training.

While end-to-end automatic speech recognition (ASR) has achieved high performance, it requires a huge amount of paired speech and transcription data for training. Recently, data augmentation methods have actively been investigated. One method is to use a text-to-speech (TTS) system to gen-erate speech data from text-only data and use the generated speech for data augmentation, but it has been found that the synthesized log Mel-scale filterbank (lmfb) features could have a serious mismatch with the real speech features. In this study, we propose a data augmentation method via a discrete speech representation. The TTS model predicts discrete ID sequences instead of lmfb features, and the ASR also uses the ID sequences as training data. We expect that the use of a discrete representation based on vq-wav2vec not only makes TTS training easier but also mitigates the mismatch with real data. Experimental evaluations show that the pro-posed method outperforms the data augmentation method using the conventional TTS. We found that it reduces speaker dependency, and the generated features are distributed more closely to the real ones.

Attention-based sequence-to-sequence (seq2seq) models have achieved promising results in automatic speech recognition (ASR).However, as these models decode in a left-to-right way, they do not have access to context on the right.We leverage both left and right context by applying BERT as an external language model to seq2seq ASR through knowledge distillation.In our proposed method, BERT generates soft labels to guide the training of seq2seq ASR.Furthermore, we leverage context beyond the current utterance as input to BERT.Experimental evaluations show that our method significantly improves the ASR performance from the seq2seq baseline on the Corpus of Spontaneous Japanese (CSJ).Knowledge distillation from BERT outperforms that from a transformer LM that only looks at left context.We also show the effectiveness of leveraging context beyond the current utterance.Our method outperforms other LM application approaches such as n-best rescoring and shallow fusion, while it does not require extra inference cost.

Monotonic chunkwise attention (MoChA) has been studied for the online streaming automatic speech recognition (ASR) based on a sequence-to-sequence framework. In contrast to connectionist temporal classification (CTC), backward probabilities cannot be leveraged in the alignment marginalization process during training due to left-to-right dependency in the decoder. This results in the error propagation of alignments to subsequent token generation. To address this problem, we propose CTC-synchronous training (CTC-ST), in which MoChA uses CTC alignments to learn optimal monotonic alignments. Reference CTC alignments are extracted from a CTC branch sharing the same encoder with the decoder. The entire model is jointly optimized so that the expected boundaries from MoChA are synchronized with the alignments. Experimental evaluations of the TEDLIUM release-2 and Librispeech corpora show that the proposed method significantly improves recognition, especially for long utterances. We also show that CTC-ST can bring out the full potential of SpecAugment for MoChA.

We investigate a monotonic multihead attention (MMA) by extending hard monotonic attention to Transformer-based automatic speech recognition (ASR) for online streaming applications. For streaming inference, all monotonic attention (MA) heads should learn proper alignments because the next token is not generated until all heads detect the corresponding token boundaries. However, we found not all MA heads learn alignments with a naïve implementation. To encourage every head to learn alignments properly, we propose HeadDrop regularization by masking out a part of heads stochastically during training. Furthermore, we propose to prune redundant heads to improve consensus among heads for boundary detection and prevent delayed token generation caused by such heads. Chunkwise attention on each MA head is extended to the multihead counterpart. Finally, we propose head-synchronous beam search decoding to guarantee stable streaming inference.

It is important to transcribe and archive speech data of endangered languages for preserving heritages of verbal culture and automatic speech recognition (ASR) is a powerful tool to facilitate this process. However, since endangered languages do not generally have large corpora with many speakers, the performance of ASR models trained on them are considerably poor in general. Nevertheless, we are often left with a lot of recordings of spontaneous speech data that have to be transcribed. In this work, for mitigating this speaker sparsity problem, we propose to convert the whole training speech data and make it sound like the test speaker in order to develop a highly accurate ASR system for this speaker. For this purpose, we utilize a CycleGAN-based non-parallel voice conversion technology to forge a labeled training data that is close to the test speaker's speech. We evaluated this speaker adaptation approach on two low-resource corpora, namely, Ainu and Mboshi. We obtained 35-60% relative improvement in phone error rate on the Ainu corpus, and 40% relative improvement was attained on the Mboshi corpus. This approach outperformed two conventional methods namely unsupervised adaptation and multilingual training with these two corpora.

Ainu is an unwritten language that has been spoken by Ainu people who are one of the ethnic groups in Japan. It is recognized as critically endangered by UNESCO and archiving and documentation of its language heritage is of paramount importance. Although a considerable amount of voice recordings of Ainu folklore has been produced and accumulated to save their culture, only a quite limited parts of them are transcribed so far. Thus, we started a project of automatic speech recognition (ASR) for the Ainu language in order to contribute to the development of annotated language archives. In this paper, we report speech corpus development and the structure and performance of end-to-end ASR for Ainu. We investigated four modeling units (phone, syllable, word piece, and word) and found that the syllable-based model performed best in terms of both word and phone recognition accuracy, which were about 60% and over 85% respectively in speaker-open condition. Furthermore, word and phone accuracy of 80% and 90% has been achieved in a speaker-closed setting. We also found out that a multilingual ASR training with additional speech corpora of English and Japanese further improves the speaker-open test accuracy.

The acoustic-to-word (A2W) automatic speech recognition (ASR) realizes very fast decoding with a simple architecture and achieves state-of-the-art performance. However, the A2W model suffers from the out-of-vocabulary (OOV) word problem and cannot use text-only data to improve the language modeling capability. Meanwhile, sequence-to-sequence neural speech synthesis has also been developed and achieved naturalness comparable to human speech. We investigate leveraging sequence-to-sequence neural speech synthesis to augment training data for the ASR system in a target domain. While speech synthesis model is usually trained with single speaker data, ASR needs to cover a variety of speakers. In this work, we extend the speech synthesizer so that it can output speech of many speakers. The multi-speaker speech synthesizer is trained with a large corpus in the source domain, then used to generate acoustic features from texts of the target domain. These synthesized speech features are combined with real speech features of the source domain to train an attention-based A2W model. Experimental results show that the A2W model trained with the multi-speaker model achieved a significant improvement over the baseline and the single speaker model.

This paper describes multichannel speech enhancement for improving automatic speech recognition (ASR) in noisy environments. Recently, the minimum variance distortionless response (MVDR) beamforming has widely been used because it works well if the steering vector of speech and the spatial covariance matrix (SCM) of noise are given. To estimating such spatial information, conventional studies take a supervised approach that classifies each time-frequency (TF) bin into noise or speech by training a deep neural network (DNN). The performance of ASR, however, is degraded in an unknown noisy environment. To solve this problem, we take an unsupervised approach that decomposes each TF bin into the sum of speech and noise by using multichannel nonnegative matrix factorization (MNMF). This enables us to accurately estimate the SCMs of speech and noise not from observed noisy mixtures but from separated speech and noise components. In this paper, we propose online MVDR beamforming by effectively initializing and incrementally updating the parameters of MNMF. Another main contribution is to comprehensively investigate the performances of ASR obtained by various types of spatial filters, i.e., time-invariant and variant versions of MVDR beamformers and those of rank-1 and full-rank multichannel Wiener filters, in combination with MNMF. The experimental results showed that the proposed method outperformed the state-of-the-art DNN-based beamforming method in unknown environments that did not match training data.

This paper presents a statistical method of single-channel speech enhancement that uses a variational autoencoder (VAE) as a prior distribution on clean speech. A standard approach to speech enhancement is to train a deep neural network (DNN) to take noisy speech as input and output clean speech. Although this supervised approach requires a very large amount of pair data for training, it is not robust against unknown environments. Another approach is to use non-negative matrix factorization (NMF) based on basis spectra trained on clean speech in advance and those adapted to noise on the fly. This semi-supervised approach, however, causes considerable signal distortion in enhanced speech due to the unrealistic assumption that speech spectrograms are linear combinations of the basis spectra. Replacing the poor linear generative model of clean speech in NMF with a VAE-a powerful nonlinear deep generative model-trained on clean speech, we formulate a unified probabilistic generative model of noisy speech. Given noisy speech as observed data, we can sample clean speech from its posterior distribution. The proposed method outperformed the conventional DNN-based method in unseen noisy environments.

Social signals such as laughter and fillers are often observed in natural conversation, and they play various roles in human-to-human communication. Detecting these events is useful for transcription systems to generate rich transcription and for dialogue systems to behave as we do such as synchronized laughing or attentive listening. We have studied an end-to-end approach to directly detect social signals from speech by using connectionist temporal classification (CTC), which is one of the end-to-end sequence labelling models. In this work, we propose a unified framework that integrates social signal detection (SSD) and automatic speech recognition (ASR). We investigate several reference labelling methods regarding social signals. Experimental evaluations demonstrate that our end-to-end framework significantly outperforms the conventional DNN-HMM system with regard to SSD performance as well as the character error rate (CER).

This paper presents unsupervised multichannel speech enhancement for noisy speech recognition. Time-frequency (TF) mask estimation has actively been studied for estimating the steering vectors and spatial covariance matrices of speech and noise used for beamforming. The state-of-the-art approach to mask estimation is to use deep neural networks (DNN s) for classifying the TF bins of observed signals into speech and noise. Such a supervised approach, however, does not work well in an unknown environment. To accurately estimate the spatial covariance matrices in an unsupervised manner, we perform blind source separation (BSS) based on multichannel nonnegative matrix factorization (MNMF) for decomposing each TF bin into the components of speech and the other sources (noise). To clarify a suitable type of beamforming for MNMF, we tested both time- invariant and time-varying versions of the minimum variance distortionless response (MVDR) beamforming in addition to standard multichannel Wiener filtering (MWF). The experimental results showed that our MNMF-based beamforming approach outperformed the state-of-the-art DNN-based beamforming method in unknown environments that do not match the training data.

This paper addresses end-to-end speech recognition which directly maps acoustic features to a word sequence. The acoustic-to-word model is attractive since it does not require an external language model and an elaborate decoder, resulting in extremely simple and fast decoding. The apparent drawback of this modeling is sparseness of training data, particularly for less frequent words. In this paper, we propose a framework complemented with a character-level model. Joint training of the word-level model with the character-level model enhances the generality of deep learning of feature extraction and classification processes, preventing it from overfitting. Moreover, the character-level model is used to decode out-of-vocabulary (OOV) words that are not covered by the word-level model. Since there are choices of connectionist temporal classification (CTC) and attention-based models in the end-to-end recognition, we also explore optimal combination for the hybrid system. Evaluations on the Corpus of Spontaneous Japanese (CSJ) show that (1) the acoustic-to-word attention-based model outperforms CTC, (2) multitask learning (MTL) with character-level CTC model is effective, and (3) the hybrid system achieves comparable or even better accuracy than the standard DNN-HMM system with a decoding speed faster by a factor of 25.

Acoustic-to-word (A2W) end-to-end automatic speech recognition (ASR) systems have attracted attention because of an extremely simplified architecture and fast decoding. To alleviate data sparseness issues due to infrequent words, the combination with an acoustic-to-character (A2C) model is investigated. Moreover, the A2C model can be used to recover-of-vocabulary (OOV) words that are not covered by the A2W model, but this requires accurate detection of OOV words. A2W models learn contexts with both acoustic and transcripts; therefore they tend to falsely recognize OOV words as words in the vocabulary. In this paper, we tackle this problem by using external language models (LM), which are trained only with transcriptions and have better linguistic information to detect OOV words. The A2C model is used to resolve these OOV words. Experimental evaluations show that external LMs have the effects of not only reducing errors but also increasing the number of detected OOV words, and the proposed method significantly improves performances in English conversational and Japanese lecture corpora, especially for-of-domain scenario. We also investigate the impact of the vocabulary size of A2W models and the data size for training LMs. Moreover, our approach can reduce the vocabulary size several times with marginal performance degradation.

Encoder-decoder models for acoustic-to-word (A2W) automatic speech recognition (ASR) are attractive for their simplicity of architecture and run-time latency while achieving state-of-the-art performances. However, word-based models commonly suffer from the-of-vocabulary (OOV) word problem. They also cannot leverage text data to improve their language modeling capability. Recently, sequence-to-sequence neural speech synthesis models trainable from corpora have been developed and shown to achieve naturalness com- parable to recorded human speech. In this paper, we explore how we can leverage the current speech synthesis technology to tailor the ASR system for a target domain by preparing only a relevant text corpus. From a set of target domain texts, we generate speech features using a sequence-to-sequence speech synthesizer. These artificial speech features together with real speech features from conventional speech corpora are used to train an attention-based A2W model. Experimental results show that the proposed approach improves the word accuracy significantly compared to the baseline trained only with the real speech, although synthetic part of the training data comes only from a single female speaker voice.

Automatic speech recognition (ASR) systems often does not perform well when it is used in a different acoustic domain from the training time, such as utterances spoken in noisy environments or in different speaking styles. We propose a novel approach to cross-domain speech recognition based on acoustic feature mappings provided by a deep neural network, which is trained using nonparallel speech corpora from two different domains and using no phone labels. For training a target domain acoustic model, we generate “fake” target speech features from the labeleld source domain features using a mapping Gf. We can also generate “fake” source features for testing from the target features using the backward mapping Gbwhich has been learned simultaneously with G f. The mappings G f and Gbare trained as adversarial networks using a conventional adversarial loss and a cycle-consistency loss criterion that encourages the backward mapping to bring the translated feature back to the original as much as possible such that Gb(Gf (x)) ≈ x. In a highly challenging task of model adaptation only using domain speech features, our method achieved up to 16 % relative improvements in WER in the evaluation using the CHiME3 real test data. The backward mapping was also confirmed to be effective with a speaking style adaptation task.

It is very important to exploit abundant unlabeled speech for improving the acoustic model training in automatic speech recognition (ASR). Semi-supervised training methods incorporate unlabeled data in addition to labeled data to enhance the model training, but it encounters the error-prone label problem. The ensemble training scheme trains a set of models and combines them to make the model more general and robust, but it has not been applied to the unlabeled data. In this work, we propose an effective semi-supervised training of deep neural network (DNN) acoustic models by incorporating the diversity among the ensemble of models. The resultant model improved the performance in the lecture transcription task. Moreover, the proposed method has also shown a potential for DNN adaptation.

This paper addresses reverberant speech recognition based on front-end processing using DAE (Deep AutoEncoder) coupled with DNN (Deep Neural Network) acoustic model. DAE can effectively and flexibly learn mapping from corrupted speech to the original clean speech based on the deep learning scheme. While this mapping is conventionally conducted only with the acoustic information, we presume the mapping is also dependent on the phone information. Therefore, we propose a new scheme (pDAE), which augments a phone-class feature to the standard acoustic features as input. Two types of the phone-class feature are investigated. One is the hard recognition result of monophones, and the other is a soft representation derived from the posterior outputs of monophone DNN. In the evaluation on the Reverb Challenge 2014 task, the augmented feature in either type results in a significant improvement (7-8% relative) from the standard DAE. It is also shown that using the soft representation in the training phase is critical.

Recently, neural networks have been used for not only phone recognition but also denoising and dereverberation. However, the conventional denoising deep autoencoder (DAE) based on the feed-forward structure is not capable of handling very long speech frames of reverberation. LSTM can be effectively trained to reduce the average error between the enhanced signal and the original clean signal by considering the effect of the long past time frames. In this paper, we demonstrate that considering as long as the maximum reverberation time of the database is effective. Since the effect of reverberation varies depending on the phone-class of the whole speech context, we augment the input of the autoencoder with the phone-class information of the past frames as well as the current frame and call this version of the LSTM autoencoder pLSTM. In the speech recognition experiment using the data set of Reverb Challenge 2014, the LSTM front-end reduced the WER of the multicondition DNN-HMM by 14.5%, and the use of the phone class feature yielded in pLSTM further improvement of 7.5%. The performance with the pLSTM is comparable to that of pDAE, while the number of parameters is only 1/25-1/8. Index Terms: Speech Dereverberation, Long Short-Term Memory (LSTM), Deep Autoencoder (DAE)

Update of acoustic and language models is vital to maintain performance of automatic speech recognition (ASR) systems. To alleviate efforts for updating models, we propose a “semi-automated ” framework for the ASR system of the Japanese National Congress. The framework consists of our speaking-style transformation (SST) and lightly-supervised training (LSV) approaches, which can automatically generate spoken-style training texts and labels from documents like meeting minutes. An experimental evaluation demonstrated that this update framework improved the ASR performance for the latest meeting data. We also address an estimation method of the ASR accuracy based on SST, which uses minutes as reference texts and does not require verbatim transcripts. Index Terms: Spontaneous speech recognition, congressional speech, lightly-supervised training, speaking-style transformation 1.

This paper presents a new approach to language model construction, learning a language model not from text, but directly from continuous speech. A phoneme lattice is created using acoustic model scores, and Bayesian techniques are used to robustly learn a language model from this noisy input. A novel sampling technique is devised that allows for the integrated learning of word boundaries and an n-gram language model with no prior linguistic knowledge. The proposed techniques were used to learn a language model directly from continuous, potentially large-vocabulary speech. This language model was able to significantly reduce the ASR phoneme error rate over a separate set of test data, and the proposed lattice processing and lexical acquisition techniques were found to be important factors in this improvement. Index Terms: language acquisition, word segmentation, Pitman-Yor language model, Bayesian learning

For effective training of acoustic and language models for spontaneous speech such as meetings, it is significant to exploit the texts available in a large scale, which may not be faithful transcripts of the utterances. We have proposed a language model transformation scheme to cope with the differences between verbatim transcripts of spontaneous utterances and human-made transcripts such as those in proceedings. In this paper, we investigate its application to lightly supervised training of the acoustic model. By transforming the corresponding text in the proceedings, we can generate a very constrained model to predict the actual utterances. The experimental evaluation with the transcription system for the Japanese Congress meetings demonstrated that the proposed scheme can generate accurate labels for acoustic model training and thus realizes the comparable ASR (Automatic Speech Recognition) performance to the case using manual transcripts.

This paper presents an automatic speech recognition (ASR) system for assisting meeting record creation of the National Congress of Japan. The system is designed to cope with spontaneous characteristics of meeting speech, as well as a variety of topics and speakers. For acoustic model, minimum phone error (MPE) training is applied with several normalization techniques. For language model, we have proposed statistical style transformation to generate spoken-style N-grams and their statistics. We also introduce statistical modeling of pronunciation variation in spontaneous speech. The ASR system was evaluated on real congressional meetings, and achieved word accuracy of 84%. It is also suggested that the ASR-based transcripts with this accuracy level is usable for editing meeting records. Index Terms: Spontaneous speech recognition, congressional speech, minimum phone error training, statistical style transformation