Listening to Multi-talker Conversations: Modular and End-to-end Perspectives

Since the first speech recognition systems were built more than 30 years ago, improvement in voice technology has enabled applications such as smart assistants and automated customer support. However, conversation intelligence of the future requires recognizing free-flowing multi-party conversations, which is a crucial and challenging component that still remains unsolved. In this dissertation, we focus on this problem of speaker-attributed multi-talker speech recognition, and propose two perspectives which result from its probabilistic formulation. In the modular perspective, we build a pipeline of sub-tasks involving speaker diarization, target speaker extraction, and speech recognition. Our first contribution is a method to perform overlap-aware diarization by reformulating spectral clustering as a constrained optimization problem. We also describe an algorithm to ensemble diarization outputs, either to combine overlap-aware systems or to perform multi-channel diarization by late fusion. Once speaker segments are identified, we robustly extract single-speaker utterances from the mixture using a GPU-accelerated implementation of guided source separation, which allows us to use an off-the-shelf ASR system to obtain speaker-attributed transcripts. Since the modular approach suffers from error propagation, we propose an alternate "end-to-end" perspective on the problem. For this, we describe the Streaming Unmixing and Recognition Transducer (SURT). We show how to train SURT models efficiently by carefully designing the network architecture, objective functions, and mixture simulation techniques. Finally, we add an auxiliary speaker branch to enable joint prediction of speaker labels synchronized with the speech tokens. We demonstrate that training on synthetic mixtures and adapting with real data helps these models transfer well for streaming transcription of real meeting sessions.

On Speaker Attribution with SURT

The Streaming Unmixing and Recognition Transducer (SURT) has recently become a popular framework for continuous, streaming, multi-talker speech recognition (ASR). With advances in architecture, objectives, and mixture simulation methods, it was demonstrated that SURT can be an efficient streaming method for speaker-agnostic transcription of real meetings. In this work, we push this framework further by proposing methods to perform speaker-attributed transcription with SURT, for both short mixtures and long recordings. We achieve this by adding an auxiliary speaker branch to SURT, and synchronizing its label prediction with ASR token prediction through HAT-style blank factorization. In order to ensure consistency in relative speaker labels across different utterance groups in a recording, we propose "speaker prefixing" -- appending each chunk with high-confidence frames of speakers identified in previous chunks, to establish the relative order. We perform extensive ablation experiments on synthetic LibriSpeech mixtures to validate our design choices, and demonstrate the efficacy of our final model on the AMI corpus.

Updated Corpora and Benchmarks for Long-Form Speech Recognition

The vast majority of ASR research uses corpora in which both the training and test data have been pre-segmented into utterances. In most real-word ASR use-cases, however, test audio is not segmented, leading to a mismatch between inference-time conditions and models trained on segmented utterances. In this paper, we re-release three standard ASR corpora - TED-LIUM 3, Gigapeech, and VoxPopuli-en - with updated transcription and alignments to enable their use for long-form ASR research. We use these reconstituted corpora to study the train-test mismatch problem for transducers and attention-based encoder-decoders (AEDs), confirming that AEDs are more susceptible to this issue. Finally, we benchmark a simple long-form training for these models, showing its efficacy for model robustness under this domain shift.

Learning from Flawed Data: Weakly Supervised Automatic Speech Recognition

Training automatic speech recognition (ASR) systems requires large amounts of well-curated paired data. However, human annotators usually perform "non-verbatim" transcription, which can result in poorly trained models. In this paper, we propose Omni-temporal Classification (OTC), a novel training criterion that explicitly incorporates label uncertainties originating from such weak supervision. This allows the model to effectively learn speech-text alignments while accommodating errors present in the training transcripts. OTC extends the conventional CTC objective for imperfect transcripts by leveraging weighted finite state transducers. Through experiments conducted on the LibriSpeech and LibriVox datasets, we demonstrate that training ASR models with OTC avoids performance degradation even with transcripts containing up to 70% errors, a scenario where CTC models fail completely. Our implementation is available at https://github.com/k2-fsa/icefall.

Training dynamic models using early exits for automatic speech recognition on resource-constrained devices

The possibility of dynamically modifying the computational load of neural models at inference time is crucial for on-device processing, where computational power is limited and time-varying. Established approaches for neural model compression exist, but they provide architecturally static models. In this paper, we investigate the use of early-exit architectures, that rely on intermediate exit branches, applied to large-vocabulary speech recognition. This allows for the development of dynamic models that adjust their computational cost to the available resources and recognition performance. Unlike previous works, besides using pre-trained backbones we also train the model from scratch with an early-exit architecture. Experiments on public datasets show that early-exit architectures from scratch not only preserve performance levels when using fewer encoder layers, but also improve task accuracy as compared to using single-exit models or using pre-trained models. Additionally, we investigate an exit selection strategy based on posterior probabilities as an alternative to frame-based entropy.

The CHiME-7 DASR Challenge: Distant Meeting Transcription with Multiple Devices in Diverse Scenarios

The CHiME challenges have played a significant role in the development and evaluation of robust automatic speech recognition (ASR) systems. We introduce the CHiME-7 distant ASR (DASR) task, within the 7th CHiME challenge. This task comprises joint ASR and diarization in far-field settings with multiple, and possibly heterogeneous, recording devices. Different from previous challenges, we evaluate systems on 3 diverse scenarios: CHiME-6, DiPCo, and Mixer 6. The goal is for participants to devise a single system that can generalize across different array geometries and use cases with no a-priori information. Another departure from earlier CHiME iterations is that participants are allowed to use open-source pre-trained models and datasets. In this paper, we describe the challenge design, motivation, and fundamental research questions in detail. We also present the baseline system, which is fully array-topology agnostic and features multi-channel diarization, channel selection, guided source separation and a robust ASR model that leverages self-supervised speech representations (SSLR).

SURT 2.0: Advances in Transducer-based Multi-talker Speech Recognition

The Streaming Unmixing and Recognition Transducer (SURT) model was proposed recently as an end-to-end approach for continuous, streaming, multi-talker speech recognition (ASR). Despite impressive results on multi-turn meetings, SURT has notable limitations: (i) it suffers from leakage and omission related errors; (ii) it is computationally expensive, due to which it has not seen adoption in academia; and (iii) it has only been evaluated on synthetic mixtures. In this work, we propose several modifications to the original SURT which are carefully designed to fix the above limitations. In particular, we (i) change the unmixing module to a mask estimator that uses dual-path modeling, (ii) use a streaming zipformer encoder and a stateless decoder for the transducer, (iii) perform mixture simulation using force-aligned subsegments, (iv) pre-train the transducer on single-speaker data, (v) use auxiliary objectives in the form of masking loss and encoder CTC loss, and (vi) perform domain adaptation for far-field recognition. We show that our modifications allow SURT 2.0 to outperform its predecessor in terms of multi-talker ASR results, while being efficient enough to train with academic resources. We conduct our evaluations on 3 publicly available meeting benchmarks -- LibriCSS, AMI, and ICSI, where our best model achieves WERs of 16.9%, 44.6% and 32.2%, respectively, on far-field unsegmented recordings. We release training recipes and pre-trained models: https://sites.google.com/view/surt2.

GPU-accelerated Guided Source Separation for Meeting Transcription

Guided source separation (GSS) is a type of target-speaker extraction method that relies on pre-computed speaker activities and blind source separation to perform front-end enhancement of overlapped speech signals. It was first proposed during the CHiME-5 challenge and provided significant improvements over the delay-and-sum beamforming baseline. Despite its strengths, however, the method has seen limited adoption for meeting transcription benchmarks primarily due to its high computation time. In this paper, we describe our improved implementation of GSS that leverages the power of modern GPU-based pipelines, including batched processing of frequencies and segments, to provide 300x speed-up over CPU-based inference. The improved inference time allows us to perform detailed ablation studies over several parameters of the GSS algorithm -- such as context duration, number of channels, and noise class, to name a few. We provide end-to-end reproducible pipelines for speaker-attributed transcription of popular meeting benchmarks: LibriCSS, AMI, and AliMeeting. Our code and recipes are publicly available: https://github.com/desh2608/gss.

Adapting self-supervised models to multi-talker speech recognition using speaker embeddings

Self-supervised learning (SSL) methods which learn representations of data without explicit supervision have gained popularity in speech-processing tasks, particularly for single-talker applications. However, these models often have degraded performance for multi-talker scenarios -- possibly due to the domain mismatch -- which severely limits their use for such applications. In this paper, we investigate the adaptation of upstream SSL models to the multi-talker automatic speech recognition (ASR) task under two conditions. First, when segmented utterances are given, we show that adding a target speaker extraction (TSE) module based on enrollment embeddings is complementary to mixture-aware pre-training. Second, for unsegmented mixtures, we propose a novel joint speaker modeling (JSM) approach, which aggregates information from all speakers in the mixture through their embeddings. With controlled experiments on Libri2Mix, we show that using speaker embeddings provides relative WER improvements of 9.1% and 42.1% over strong baselines for the segmented and unsegmented cases, respectively. We also demonstrate the effectiveness of our models for real conversational mixtures through experiments on the AMI dataset.

Leveraging Speech Separation for Conversational Telephone Speaker Diarization

Speech separation and speaker diarization have strong similarities. In particular with respect to end-to-end neural diarization (EEND) methods. Separation aims at extracting each speaker from overlapped speech, while diarization identifies time boundaries of speech segments produced by the same speaker. In this paper, we carry out an analysis of the use of speech separation guided diarization (SSGD) where diarization is performed simply by separating the speakers signals and applying voice activity detection. In particular we compare two speech separation (SSep) models, both in offline and online settings. In the online setting we consider both the use of continuous source separation (CSS) and causal SSep models architectures. As an additional contribution, we show a simple post-processing algorithm which reduces significantly the false alarm errors of a SSGD pipeline. We perform our experiments on Fisher Corpus Part 1 and CALLHOME datasets evaluating both separation and diarization metrics. Notably, without fine-tuning, our SSGD DPRNN-based online model achieves 12.7% DER on CALLHOME, comparable with state-of-the-art EEND models despite having considerably lower latency, i.e., 50 ms vs 1 s.