Yanzhang He

Shuo-yiin Chang, Bo Li, Tara N. Sainath et al.

While a streaming voice assistant system has been used in many applications, this system typically focuses on unnatural, one-shot interactions assuming input from a single voice query without hesitation or disfluency. However, a common conversational utterance often involves multiple queries with turn-taking, in addition to disfluencies. These disfluencies include pausing to think, hesitations, word lengthening, filled pauses and repeated phrases. This makes doing speech recognition with conversational speech, including one with multiple queries, a challenging task. To better model the conversational interaction, it is critical to discriminate disfluencies and end of query in order to allow the user to hold the floor for disfluencies while having the system respond as quickly as possible when the user has finished speaking. In this paper, we present a turntaking predictor built on top of the end-to-end (E2E) speech recognizer. Our best system is obtained by jointly optimizing for ASR task and detecting when the user is paused to think or finished speaking. The proposed approach demonstrates over 97% recall rate and 85% precision rate on predicting true turn-taking with only 100 ms latency on a test set designed with 4 types of disfluencies inserted in conversational utterances.

Shaojin Ding, Weiran Wang, Ding Zhao et al.

In this paper, we propose a dynamic cascaded encoder Automatic Speech Recognition (ASR) model, which unifies models for different deployment scenarios. Moreover, the model can significantly reduce model size and power consumption without loss of quality. Namely, with the dynamic cascaded encoder model, we explore three techniques to maximally boost the performance of each model size: 1) Use separate decoders for each sub-model while sharing the encoders; 2) Use funnel-pooling to improve the encoder efficiency; 3) Balance the size of causal and non-causal encoders to improve quality and fit deployment constraints. Overall, the proposed large-medium model has 30% smaller size and reduces power consumption by 33%, compared to the baseline cascaded encoder model. The triple-size model that unifies the large, medium, and small models achieves 37% total size reduction with minimal quality loss, while substantially reducing the engineering efforts of having separate models.

Bo Li, Tara N. Sainath, Ruoming Pang et al.

On-device end-to-end (E2E) models have shown improvements over a conventional model on English Voice Search tasks in both quality and latency. E2E models have also shown promising results for multilingual automatic speech recognition (ASR). In this paper, we extend our previous capacity solution to streaming applications and present a streaming multilingual E2E ASR system that runs fully on device with comparable quality and latency to individual monolingual models. To achieve that, we propose an Encoder Endpointer model and an End-of-Utterance (EOU) Joint Layer for a better quality and latency trade-off. Our system is built in a language agnostic manner allowing it to natively support intersentential code switching in real time. To address the feasibility concerns on large models, we conducted on-device profiling and replaced the time consuming LSTM decoder with the recently developed Embedding decoder. With these changes, we managed to run such a system on a mobile device in less than real time.

Weiran Wang, Tongzhou Chen, Tara N. Sainath et al.

Language models (LMs) significantly improve the recognition accuracy of end-to-end (E2E) models on words rarely seen during training, when used in either the shallow fusion or the rescoring setups. In this work, we introduce LMs in the learning of hybrid autoregressive transducer (HAT) models in the discriminative training framework, to mitigate the training versus inference gap regarding the use of LMs. For the shallow fusion setup, we use LMs during both hypotheses generation and loss computation, and the LM-aware MWER-trained model achieves 10\% relative improvement over the model trained with standard MWER on voice search test sets containing rare words. For the rescoring setup, we learn a small neural module to generate per-token fusion weights in a data-dependent manner. This model achieves the same rescoring WER as regular MWER-trained model, but without the need for sweeping fusion weights.

Steven M. Hernandez, Ding Zhao, Shaojin Ding et al.

Continued improvements in machine learning techniques offer exciting new opportunities through the use of larger models and larger training datasets. However, there is a growing need to offer these new capabilities on-board low-powered devices such as smartphones, wearables and other embedded environments where only low memory is available. Towards this, we consider methods to reduce the model size of Conformer-based speech recognition models which typically require models with greater than 100M parameters down to just $5$M parameters while minimizing impact on model quality. Such a model allows us to achieve always-on ambient speech recognition on edge devices with low-memory neural processors. We propose model weight reuse at different levels within our model architecture: (i) repeating full conformer block layers, (ii) sharing specific conformer modules across layers, (iii) sharing sub-components per conformer module, and (iv) sharing decomposed sub-component weights after low-rank decomposition. By sharing weights at different levels of our model, we can retain the full model in-memory while increasing the number of virtual transformations applied to the input. Through a series of ablation studies and evaluations, we find that with weight sharing and a low-rank architecture, we can achieve a WER of 2.84 and 2.94 for Librispeech dev-clean and test-clean respectively with a $5$M parameter model.

Ke Hu, Tara N. Sainath, Yanzhang He et al.

Text-only and semi-supervised training based on audio-only data has gained popularity recently due to the wide availability of unlabeled text and speech data. In this work, we propose incorporating text-only and semi-supervised training into an attention-based deliberation model. By incorporating text-only data in training a bidirectional encoder representation from transformer (BERT) for the deliberation text encoder, and large-scale text-to-speech and audio-only utterances using joint acoustic and text decoder (JATD) and semi-supervised training, we achieved 4%-12% WER reduction for various tasks compared to the baseline deliberation. Compared to a state-of-the-art language model (LM) rescoring method, the deliberation model reduces the Google Voice Search WER by 11% relative. We show that the deliberation model also achieves a positive human side-by-side evaluation compared to the state-of-the-art LM rescorer with reasonable endpointer latencies.

Shaan Bijwadia, Shuo-yiin Chang, Bo Li et al.

Automatic speech recognition (ASR) systems typically rely on an external endpointer (EP) model to identify speech boundaries. In this work, we propose a method to jointly train the ASR and EP tasks in a single end-to-end (E2E) multitask model, improving EP quality by optionally leveraging information from the ASR audio encoder. We introduce a "switch" connection, which trains the EP to consume either the audio frames directly or low-level latent representations from the ASR model. This results in a single E2E model that can be used during inference to perform frame filtering at low cost, and also make high quality end-of-query (EOQ) predictions based on ongoing ASR computation. We present results on a voice search test set showing that, compared to separate single-task models, this approach reduces median endpoint latency by 120 ms (30.8% reduction), and 90th percentile latency by 170 ms (23.0% reduction), without regressing word error rate. For continuous recognition, WER improves by 10.6% (relative).

W. Ronny Huang, Shuo-Yiin Chang, Tara N. Sainath et al.

We explore unifying a neural segmenter with two-pass cascaded encoder ASR into a single model. A key challenge is allowing the segmenter (which runs in real-time, synchronously with the decoder) to finalize the 2nd pass (which runs 900 ms behind real-time) without introducing user-perceived latency or deletion errors during inference. We propose a design where the neural segmenter is integrated with the causal 1st pass decoder to emit a end-of-segment (EOS) signal in real-time. The EOS signal is then used to finalize the non-causal 2nd pass. We experiment with different ways to finalize the 2nd pass, and find that a novel dummy frame injection strategy allows for simultaneous high quality 2nd pass results and low finalization latency. On a real-world long-form captioning task (YouTube), we achieve 2.4% relative WER and 140 ms EOS latency gains over a baseline VAD-based segmenter with the same cascaded encoder.

Shaojin Ding, Rajeev Rikhye, Qiao Liang et al.

Personalization of on-device speech recognition (ASR) has seen explosive growth in recent years, largely due to the increasing popularity of personal assistant features on mobile devices and smart home speakers. In this work, we present Personal VAD 2.0, a personalized voice activity detector that detects the voice activity of a target speaker, as part of a streaming on-device ASR system. Although previous proof-of-concept studies have validated the effectiveness of Personal VAD, there are still several critical challenges to address before this model can be used in production: first, the quality must be satisfactory in both enrollment and enrollment-less scenarios; second, it should operate in a streaming fashion; and finally, the model size should be small enough to fit a limited latency and CPU/Memory budget. To meet the multi-faceted requirements, we propose a series of novel designs: 1) advanced speaker embedding modulation methods; 2) a new training paradigm to generalize to enrollment-less conditions; 3) architecture and runtime optimizations for latency and resource restrictions. Extensive experiments on a realistic speech recognition system demonstrated the state-of-the-art performance of our proposed method.

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Rohit Prabhavalkar, Zhong Meng, Weiran Wang et al.

The accuracy of end-to-end (E2E) automatic speech recognition (ASR) models continues to improve as they are scaled to larger sizes, with some now reaching billions of parameters. Widespread deployment and adoption of these models, however, requires computationally efficient strategies for decoding. In the present work, we study one such strategy: applying multiple frame reduction layers in the encoder to compress encoder outputs into a small number of output frames. While similar techniques have been investigated in previous work, we achieve dramatically more reduction than has previously been demonstrated through the use of multiple funnel reduction layers. Through ablations, we study the impact of various architectural choices in the encoder to identify the most effective strategies. We demonstrate that we can generate one encoder output frame for every 2.56 sec of input speech, without significantly affecting word error rate on a large-scale voice search task, while improving encoder and decoder latencies by 48% and 92% respectively, relative to a strong but computationally expensive baseline.

Antoine Bruguier, David Qiu, Yanzhang He

For many streaming automatic speech recognition tasks, it is important to provide timely intermediate streaming results, while refining a high quality final result. This can be done using a multi-stage architecture, where a small left-context only model creates streaming results and a larger left- and right-context model produces a final result at the end. While this significantly improves the quality of the final results without compromising the streaming emission latency of the system, streaming results do not benefit from the quality improvements. Here, we propose using a text manipulation algorithm that merges the streaming outputs of both models. We improve the quality of streaming results by around 10%, without altering the final results. Our approach introduces no additional latency and reduces flickering. It is also lightweight, does not require retraining the model, and it can be applied to a wide variety of multi-stage architectures.

David Qiu, David Rim, Shaojin Ding et al.

With the rapid increase in the size of neural networks, model compression has become an important area of research. Quantization is an effective technique at decreasing the model size, memory access, and compute load of large models. Despite recent advances in quantization aware training (QAT) technique, most papers present evaluations that are focused on computer vision tasks, which have different training dynamics compared to sequence tasks. In this paper, we first benchmark the impact of popular techniques such as straight through estimator, pseudo-quantization noise, learnable scale parameter, clipping, etc. on 4-bit seq2seq models across a suite of speech recognition datasets ranging from 1,000 hours to 1 million hours, as well as one machine translation dataset to illustrate its applicability outside of speech. Through the experiments, we report that noise based QAT suffers when there is insufficient regularization signal flowing back to the quantization scale. We propose low complexity changes to the QAT process to improve model accuracy (outperforming popular learnable scale and clipping methods). With the improved accuracy, it opens up the possibility to exploit some of the other benefits of noise based QAT: 1) training a single model that performs well in mixed precision mode and 2) improved generalization on long form speech recognition.

Shaojin Ding, Phoenix Meadowlark, Yanzhang He et al.

Reducing the latency and model size has always been a significant research problem for live Automatic Speech Recognition (ASR) application scenarios. Along this direction, model quantization has become an increasingly popular approach to compress neural networks and reduce computation cost. Most of the existing practical ASR systems apply post-training 8-bit quantization. To achieve a higher compression rate without introducing additional performance regression, in this study, we propose to develop 4-bit ASR models with native quantization aware training, which leverages native integer operations to effectively optimize both training and inference. We conducted two experiments on state-of-the-art Conformer-based ASR models to evaluate our proposed quantization technique. First, we explored the impact of different precisions for both weight and activation quantization on the LibriSpeech dataset, and obtained a lossless 4-bit Conformer model with 5.8x size reduction compared to the float32 model. Following this, we for the first time investigated and revealed the viability of 4-bit quantization on a practical ASR system that is trained with large-scale datasets, and produced a lossless Conformer ASR model with mixed 4-bit and 8-bit weights that has 5x size reduction compared to the float32 model.

Rajeev Rikhye, Quan Wang, Qiao Liang et al.

VoiceFilter-Lite is a speaker-conditioned voice separation model that plays a crucial role in improving speech recognition and speaker verification by suppressing overlapping speech from non-target speakers. However, one limitation of VoiceFilter-Lite, and other speaker-conditioned speech models in general, is that these models are usually limited to a single target speaker. This is undesirable as most smart home devices now support multiple enrolled users. In order to extend the benefits of personalization to multiple users, we previously developed an attention-based speaker selection mechanism and applied it to VoiceFilter-Lite. However, the original multi-user VoiceFilter-Lite model suffers from significant performance degradation compared with single-user models. In this paper, we devised a series of experiments to improve the multi-user VoiceFilter-Lite model. By incorporating a dual learning rate schedule and by using feature-wise linear modulation (FiLM) to condition the model with the attended speaker embedding, we successfully closed the performance gap between multi-user and single-user VoiceFilter-Lite models on single-speaker evaluations. At the same time, the new model can also be easily extended to support any number of users, and significantly outperforms our previously published model on multi-speaker evaluations.

Arun Narayanan, Chung-Cheng Chiu, Tom O'Malley et al.

This work introduces \emph{cross-attention conformer}, an attention-based architecture for context modeling in speech enhancement. Given that the context information can often be sequential, and of different length as the audio that is to be enhanced, we make use of cross-attention to summarize and merge contextual information with input features. Building upon the recently proposed conformer model that uses self attention layers as building blocks, the proposed cross-attention conformer can be used to build deep contextual models. As a concrete example, we show how noise context, i.e., short noise-only audio segment preceding an utterance, can be used to build a speech enhancement feature frontend using cross-attention conformer layers for improving noise robustness of automatic speech recognition.

Qiujia Li, Yu Zhang, David Qiu et al.

As end-to-end automatic speech recognition (ASR) models reach promising performance, various downstream tasks rely on good confidence estimators for these systems. Recent research has shown that model-based confidence estimators have a significant advantage over using the output softmax probabilities. If the input data to the speech recogniser is from mismatched acoustic and linguistic conditions, the ASR performance and the corresponding confidence estimators may exhibit severe degradation. Since confidence models are often trained on the same in-domain data as the ASR, generalising to out-of-domain (OOD) scenarios is challenging. By keeping the ASR model untouched, this paper proposes two approaches to improve the model-based confidence estimators on OOD data: using pseudo transcriptions and an additional OOD language model. With an ASR model trained on LibriSpeech, experiments show that the proposed methods can greatly improve the confidence metrics on TED-LIUM and Switchboard datasets while preserving in-domain performance. Furthermore, the improved confidence estimators are better calibrated on OOD data and can provide a much more reliable criterion for data selection.

Dongseong Hwang, Ananya Misra, Zhouyuan Huo et al.

Self- and semi-supervised learning methods have been actively investigated to reduce labeled training data or enhance the model performance. However, the approach mostly focus on in-domain performance for public datasets. In this study, we utilize the combination of self- and semi-supervised learning methods to solve unseen domain adaptation problem in a large-scale production setting for online ASR model. This approach demonstrates that using the source domain data with a small fraction of the target domain data (3%) can recover the performance gap compared to a full data baseline: relative 13.5% WER improvement for target domain data.

Rami Botros, Tara N. Sainath, Robert David et al.

Previous works on the Recurrent Neural Network-Transducer (RNN-T) models have shown that, under some conditions, it is possible to simplify its prediction network with little or no loss in recognition accuracy (arXiv:2003.07705 [eess.AS], [2], arXiv:2012.06749 [cs.CL]). This is done by limiting the context size of previous labels and/or using a simpler architecture for its layers instead of LSTMs. The benefits of such changes include reduction in model size, faster inference and power savings, which are all useful for on-device applications. In this work, we study ways to make the RNN-T decoder (prediction network + joint network) smaller and faster without degradation in recognition performance. Our prediction network performs a simple weighted averaging of the input embeddings, and shares its embedding matrix weights with the joint network's output layer (a.k.a. weight tying, commonly used in language modeling arXiv:1611.01462 [cs.LG]). This simple design, when used in conjunction with additional Edit-based Minimum Bayes Risk (EMBR) training, reduces the RNN-T Decoder from 23M parameters to just 2M, without affecting word-error rate (WER).

Rajeev Rikhye, Quan Wang, Qiao Liang et al.

In this paper, we propose a solution to allow speaker conditioned speech models, such as VoiceFilter-Lite, to support an arbitrary number of enrolled users in a single pass. This is achieved by using an attention mechanism on multiple speaker embeddings to compute a single attentive embedding, which is then used as a side input to the model. We implemented multi-user VoiceFilter-Lite and evaluated it for three tasks: (1) a streaming automatic speech recognition (ASR) task; (2) a text-independent speaker verification task; and (3) a personalized keyphrase detection task, where ASR has to detect keyphrases from multiple enrolled users in a noisy environment. Our experiments show that, with up to four enrolled users, multi-user VoiceFilter-Lite is able to significantly reduce speech recognition and speaker verification errors when there is overlapping speech, without affecting performance under other acoustic conditions. This attentive speaker embedding approach can also be easily applied to other speaker-conditioned models such as personal VAD and personalized ASR.

Rajeev Rikhye, Quan Wang, Qiao Liang et al.

In this paper, we introduce a streaming keyphrase detection system that can be easily customized to accurately detect any phrase composed of words from a large vocabulary. The system is implemented with an end-to-end trained automatic speech recognition (ASR) model and a text-independent speaker verification model. To address the challenge of detecting these keyphrases under various noisy conditions, a speaker separation model is added to the feature frontend of the speaker verification model, and an adaptive noise cancellation (ANC) algorithm is included to exploit cross-microphone noise coherence. Our experiments show that the text-independent speaker verification model largely reduces the false triggering rate of the keyphrase detection, while the speaker separation model and adaptive noise cancellation largely reduce false rejections.

David Qiu, Yanzhang He, Qiujia Li et al.

Confidence scores are very useful for downstream applications of automatic speech recognition (ASR) systems. Recent works have proposed using neural networks to learn word or utterance confidence scores for end-to-end ASR. In those studies, word confidence by itself does not model deletions, and utterance confidence does not take advantage of word-level training signals. This paper proposes to jointly learn word confidence, word deletion, and utterance confidence. Empirical results show that multi-task learning with all three objectives improves confidence metrics (NCE, AUC, RMSE) without the need for increasing the model size of the confidence estimation module. Using the utterance-level confidence for rescoring also decreases the word error rates on Google's Voice Search and Long-tail Maps datasets by 3-5% relative, without needing a dedicated neural rescorer.

David Qiu, Qiujia Li, Yanzhang He et al.

We study the problem of word-level confidence estimation in subword-based end-to-end (E2E) models for automatic speech recognition (ASR). Although prior works have proposed training auxiliary confidence models for ASR systems, they do not extend naturally to systems that operate on word-pieces (WP) as their vocabulary. In particular, ground truth WP correctness labels are needed for training confidence models, but the non-unique tokenization from word to WP causes inaccurate labels to be generated. This paper proposes and studies two confidence models of increasing complexity to solve this problem. The final model uses self-attention to directly learn word-level confidence without needing subword tokenization, and exploits full context features from multiple hypotheses to improve confidence accuracy. Experiments on Voice Search and long-tail test sets show standard metrics (e.g., NCE, AUC, RMSE) improving substantially. The proposed confidence module also enables a model selection approach to combine an on-device E2E model with a hybrid model on the server to address the rare word recognition problem for the E2E model.

Rohit Prabhavalkar, Yanzhang He, David Rybach et al.

End-to-end models that condition the output label sequence on all previously predicted labels have emerged as popular alternatives to conventional systems for automatic speech recognition (ASR). Since unique label histories correspond to distinct models states, such models are decoded using an approximate beam-search process which produces a tree of hypotheses. In this work, we study the influence of the amount of label context on the model's accuracy, and its impact on the efficiency of the decoding process. We find that we can limit the context of the recurrent neural network transducer (RNN-T) during training to just four previous word-piece labels, without degrading word error rate (WER) relative to the full-context baseline. Limiting context also provides opportunities to improve the efficiency of the beam-search process during decoding by removing redundant paths from the active beam, and instead retaining them in the final lattice. This path-merging scheme can also be applied when decoding the baseline full-context model through an approximation. Overall, we find that the proposed path-merging scheme is extremely effective allowing us to improve oracle WERs by up to 36% over the baseline, while simultaneously reducing the number of model evaluations by up to 5.3% without any degradation in WER.

Bo Li, Anmol Gulati, Jiahui Yu et al.

End-to-end (E2E) models have shown to outperform state-of-the-art conventional models for streaming speech recognition [1] across many dimensions, including quality (as measured by word error rate (WER)) and endpointer latency [2]. However, the model still tends to delay the predictions towards the end and thus has much higher partial latency compared to a conventional ASR model. To address this issue, we look at encouraging the E2E model to emit words early, through an algorithm called FastEmit [3]. Naturally, improving on latency results in a quality degradation. To address this, we explore replacing the LSTM layers in the encoder of our E2E model with Conformer layers [4], which has shown good improvements for ASR. Secondly, we also explore running a 2nd-pass beam search to improve quality. In order to ensure the 2nd-pass completes quickly, we explore non-causal Conformer layers that feed into the same 1st-pass RNN-T decoder, an algorithm called Cascaded Encoders [5]. Overall, we find that the Conformer RNN-T with Cascaded Encoders offers a better quality and latency tradeoff for streaming ASR.

Qiujia Li, David Qiu, Yu Zhang et al.

For various speech-related tasks, confidence scores from a speech recogniser are a useful measure to assess the quality of transcriptions. In traditional hidden Markov model-based automatic speech recognition (ASR) systems, confidence scores can be reliably obtained from word posteriors in decoding lattices. However, for an ASR system with an auto-regressive decoder, such as an attention-based sequence-to-sequence model, computing word posteriors is difficult. An obvious alternative is to use the decoder softmax probability as the model confidence. In this paper, we first examine how some commonly used regularisation methods influence the softmax-based confidence scores and study the overconfident behaviour of end-to-end models. Then we propose a lightweight and effective approach named confidence estimation module (CEM) on top of an existing end-to-end ASR model. Experiments on LibriSpeech show that CEM can mitigate the overconfidence problem and can produce more reliable confidence scores with and without shallow fusion of a language model. Further analysis shows that CEM generalises well to speech from a moderately mismatched domain and can potentially improve downstream tasks such as semi-supervised learning.

Jiahui Yu, Chung-Cheng Chiu, Bo Li et al.

Streaming automatic speech recognition (ASR) aims to emit each hypothesized word as quickly and accurately as possible. However, emitting fast without degrading quality, as measured by word error rate (WER), is highly challenging. Existing approaches including Early and Late Penalties and Constrained Alignments penalize emission delay by manipulating per-token or per-frame probability prediction in sequence transducer models. While being successful in reducing delay, these approaches suffer from significant accuracy regression and also require additional word alignment information from an existing model. In this work, we propose a sequence-level emission regularization method, named FastEmit, that applies latency regularization directly on per-sequence probability in training transducer models, and does not require any alignment. We demonstrate that FastEmit is more suitable to the sequence-level optimization of transducer models for streaming ASR by applying it on various end-to-end streaming ASR networks including RNN-Transducer, Transformer-Transducer, ConvNet-Transducer and Conformer-Transducer. We achieve 150-300 ms latency reduction with significantly better accuracy over previous techniques on a Voice Search test set. FastEmit also improves streaming ASR accuracy from 4.4%/8.9% to 3.1%/7.5% WER, meanwhile reduces 90th percentile latency from 210 ms to only 30 ms on LibriSpeech.

Quan Wang, Ignacio Lopez Moreno, Mert Saglam et al.

We introduce VoiceFilter-Lite, a single-channel source separation model that runs on the device to preserve only the speech signals from a target user, as part of a streaming speech recognition system. Delivering such a model presents numerous challenges: It should improve the performance when the input signal consists of overlapped speech, and must not hurt the speech recognition performance under all other acoustic conditions. Besides, this model must be tiny, fast, and perform inference in a streaming fashion, in order to have minimal impact on CPU, memory, battery and latency. We propose novel techniques to meet these multi-faceted requirements, including using a new asymmetric loss, and adopting adaptive runtime suppression strength. We also show that such a model can be quantized as a 8-bit integer model and run in realtime.

Wei Li, James Qin, Chung-Cheng Chiu et al.

Recent advances of end-to-end models have outperformed conventional models through employing a two-pass model. The two-pass model provides better speed-quality trade-offs for on-device speech recognition, where a 1st-pass model generates hypotheses in a streaming fashion, and a 2nd-pass model re-scores the hypotheses with full audio sequence context. The 2nd-pass model plays a key role in the quality improvement of the end-to-end model to surpass the conventional model. One main challenge of the two-pass model is the computation latency introduced by the 2nd-pass model. Specifically, the original design of the two-pass model uses LSTMs for the 2nd-pass model, which are subject to long latency as they are constrained by the recurrent nature and have to run inference sequentially. In this work we explore replacing the LSTM layers in the 2nd-pass rescorer with Transformer layers, which can process the entire hypothesis sequences in parallel and can therefore utilize the on-device computation resources more efficiently. Compared with an LSTM-based baseline, our proposed Transformer rescorer achieves more than 50% latency reduction with quality improvement.

Yuan Shangguan, Kate Knister, Yanzhang He et al.

The demand for fast and accurate incremental speech recognition increases as the applications of automatic speech recognition (ASR) proliferate. Incremental speech recognizers output chunks of partially recognized words while the user is still talking. Partial results can be revised before the ASR finalizes its hypothesis, causing instability issues. We analyze the quality and stability of on-device streaming end-to-end (E2E) ASR models. We first introduce a novel set of metrics that quantify the instability at word and segment levels. We study the impact of several model training techniques that improve E2E model qualities but degrade model stability. We categorize the causes of instability and explore various solutions to mitigate them in a streaming E2E ASR system. Index Terms: ASR, stability, end-to-end, text normalization,on-device, RNN-T

Tara N. Sainath, Yanzhang He, Bo Li et al.

Thus far, end-to-end (E2E) models have not been shown to outperform state-of-the-art conventional models with respect to both quality, i.e., word error rate (WER), and latency, i.e., the time the hypothesis is finalized after the user stops speaking. In this paper, we develop a first-pass Recurrent Neural Network Transducer (RNN-T) model and a second-pass Listen, Attend, Spell (LAS) rescorer that surpasses a conventional model in both quality and latency. On the quality side, we incorporate a large number of utterances across varied domains to increase acoustic diversity and the vocabulary seen by the model. We also train with accented English speech to make the model more robust to different pronunciations. In addition, given the increased amount of training data, we explore a varied learning rate schedule. On the latency front, we explore using the end-of-sentence decision emitted by the RNN-T model to close the microphone, and also introduce various optimizations to improve the speed of LAS rescoring. Overall, we find that RNN-T+LAS offers a better WER and latency tradeoff compared to a conventional model. For example, for the same latency, RNN-T+LAS obtains a 8% relative improvement in WER, while being more than 400-times smaller in model size.

Tara N. Sainath, Ruoming Pang, David Rybach et al.

The requirements for many applications of state-of-the-art speech recognition systems include not only low word error rate (WER) but also low latency. Specifically, for many use-cases, the system must be able to decode utterances in a streaming fashion and faster than real-time. Recently, a streaming recurrent neural network transducer (RNN-T) end-to-end (E2E) model has shown to be a good candidate for on-device speech recognition, with improved WER and latency metrics compared to conventional on-device models [1]. However, this model still lags behind a large state-of-the-art conventional model in quality [2]. On the other hand, a non-streaming E2E Listen, Attend and Spell (LAS) model has shown comparable quality to large conventional models [3]. This work aims to bring the quality of an E2E streaming model closer to that of a conventional system by incorporating a LAS network as a second-pass component, while still abiding by latency constraints. Our proposed two-pass model achieves a 17%-22% relative reduction in WER compared to RNN-T alone and increases latency by a small fraction over RNN-T.

Jonathan Shen, Patrick Nguyen, Yonghui Wu et al.

Lingvo is a Tensorflow framework offering a complete solution for collaborative deep learning research, with a particular focus towards sequence-to-sequence models. Lingvo models are composed of modular building blocks that are flexible and easily extensible, and experiment configurations are centralized and highly customizable. Distributed training and quantized inference are supported directly within the framework, and it contains existing implementations of a large number of utilities, helper functions, and the newest research ideas. Lingvo has been used in collaboration by dozens of researchers in more than 20 papers over the last two years. This document outlines the underlying design of Lingvo and serves as an introduction to the various pieces of the framework, while also offering examples of advanced features that showcase the capabilities of the framework.

Yanzhang He, Tara N. Sainath, Rohit Prabhavalkar et al.

End-to-end (E2E) models, which directly predict output character sequences given input speech, are good candidates for on-device speech recognition. E2E models, however, present numerous challenges: In order to be truly useful, such models must decode speech utterances in a streaming fashion, in real time; they must be robust to the long tail of use cases; they must be able to leverage user-specific context (e.g., contact lists); and above all, they must be extremely accurate. In this work, we describe our efforts at building an E2E speech recognizer using a recurrent neural network transducer. In experimental evaluations, we find that the proposed approach can outperform a conventional CTC-based model in terms of both latency and accuracy in a number of evaluation categories.

Yanzhang He, Rohit Prabhavalkar, Kanishka Rao et al.

We develop streaming keyword spotting systems using a recurrent neural network transducer (RNN-T) model: an all-neural, end-to-end trained, sequence-to-sequence model which jointly learns acoustic and language model components. Our models are trained to predict either phonemes or graphemes as subword units, thus allowing us to detect arbitrary keyword phrases, without any out-of-vocabulary words. In order to adapt the models to the requirements of keyword spotting, we propose a novel technique which biases the RNN-T system towards a specific keyword of interest. Our systems are compared against a strong sequence-trained, connectionist temporal classification (CTC) based "keyword-filler" baseline, which is augmented with a separate phoneme language model. Overall, our RNN-T system with the proposed biasing technique significantly improves performance over the baseline system.