CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models

In our previous work, we introduced CosyVoice, a multilingual speech synthesis model based on supervised discrete speech tokens. By employing progressive semantic decoding with two popular generative models, language models (LMs) and Flow Matching, CosyVoice demonstrated high prosody naturalness, content consistency, and speaker similarity in speech in-context learning. Recently, significant progress has been made in multi-modal large language models (LLMs), where the response latency and real-time factor of speech synthesis play a crucial role in the interactive experience. Therefore, in this report, we present an improved streaming speech synthesis model, CosyVoice 2, which incorporates comprehensive and systematic optimizations. Specifically, we introduce finite-scalar quantization to improve the codebook utilization of speech tokens. For the text-speech LM, we streamline the model architecture to allow direct use of a pre-trained LLM as the backbone. In addition, we develop a chunk-aware causal flow matching model to support various synthesis scenarios, enabling both streaming and non-streaming synthesis within a single model. By training on a large-scale multilingual dataset, CosyVoice 2 achieves human-parity naturalness, minimal response latency, and virtually lossless synthesis quality in the streaming mode. We invite readers to listen to the demos at https://funaudiollm.github.io/cosyvoice2.

Enhancing Low-Resource ASR through Versatile TTS: Bridging the Data Gap

While automatic speech recognition (ASR) systems have achieved remarkable performance with large-scale datasets, their efficacy remains inadequate in low-resource settings, encompassing dialects, accents, minority languages, and long-tail hotwords, domains with significant practical relevance. With the advent of versatile and powerful text-to-speech (TTS) models, capable of generating speech with human-level naturalness, expressiveness, and diverse speaker profiles, leveraging TTS for ASR data augmentation provides a cost-effective and practical approach to enhancing ASR performance. Comprehensive experiments on an unprecedentedly rich variety of low-resource datasets demonstrate consistent and substantial performance improvements, proving that the proposed method of enhancing low-resource ASR through a versatile TTS model is highly effective and has broad application prospects. Furthermore, we delve deeper into key characteristics of synthesized speech data that contribute to ASR improvement, examining factors such as text diversity, speaker diversity, and the volume of synthesized data, with text diversity being studied for the first time in this work. We hope our findings provide helpful guidance and reference for the practical application of TTS-based data augmentation and push the advancement of low-resource ASR one step further.

IntrinsicVoice: Empowering LLMs with Intrinsic Real-time Voice Interaction Abilities

Current methods of building LLMs with voice interaction capabilities rely heavily on explicit text autoregressive generation before or during speech response generation to maintain content quality, which unfortunately brings computational overhead and increases latency in multi-turn interactions. To address this, we introduce IntrinsicVoic,e an LLM designed with intrinsic real-time voice interaction capabilities. IntrinsicVoice aims to facilitate the transfer of textual capabilities of pre-trained LLMs to the speech modality by mitigating the modality gap between text and speech. Our novelty architecture, GroupFormer, can reduce speech sequences to lengths comparable to text sequences while generating high-quality audio, significantly reducing the length difference between speech and text, speeding up inference, and alleviating long-text modeling issues. Additionally, we construct a multi-turn speech-to-speech dialogue dataset named \method-500k which includes nearly 500k turns of speech-to-speech dialogues, and a cross-modality training strategy to enhance the semantic alignment between speech and text. Experimental results demonstrate that IntrinsicVoice can generate high-quality speech response with latency lower than 100ms in multi-turn dialogue scenarios. Demos are available at https://instrinsicvoice.github.io/.

CosyVoice: A Scalable Multilingual Zero-shot Text-to-speech Synthesizer based on Supervised Semantic Tokens

Recent years have witnessed a trend that large language model (LLM) based text-to-speech (TTS) emerges into the mainstream due to their high naturalness and zero-shot capacity. In this paradigm, speech signals are discretized into token sequences, which are modeled by an LLM with text as prompts and reconstructed by a token-based vocoder to waveforms. Obviously, speech tokens play a critical role in LLM-based TTS models. Current speech tokens are learned in an unsupervised manner, which lacks explicit semantic information and alignment to the text. In this paper, we propose to represent speech with supervised semantic tokens, which are derived from a multilingual speech recognition model by inserting vector quantization into the encoder. Based on the tokens, we further propose a scalable zero-shot TTS synthesizer, CosyVoice, which consists of an LLM for text-to-token generation and a conditional flow matching model for token-to-speech synthesis. Experimental results show that supervised semantic tokens significantly outperform existing unsupervised tokens in terms of content consistency and speaker similarity for zero-shot voice cloning. Moreover, we find that utilizing large-scale data further improves the synthesis performance, indicating the scalable capacity of CosyVoice. To the best of our knowledge, this is the first attempt to involve supervised speech tokens into TTS models.

An Embarrassingly Simple Approach for LLM with Strong ASR Capacity

In this paper, we focus on solving one of the most important tasks in the field of speech processing, i.e., automatic speech recognition (ASR), with speech foundation encoders and large language models (LLM). Recent works have complex designs such as compressing the output temporally for the speech encoder, tackling modal alignment for the projector, and utilizing parameter-efficient fine-tuning for the LLM. We found that delicate designs are not necessary, while an embarrassingly simple composition of off-the-shelf speech encoder, LLM, and the only trainable linear projector is competent for the ASR task. To be more specific, we benchmark and explore various combinations of LLMs and speech encoders, leading to the optimal LLM-based ASR system, which we call SLAM-ASR. The proposed SLAM-ASR provides a clean setup and little task-specific design, where only the linear projector is trained. To the best of our knowledge, SLAM-ASR achieves the best performance on the Librispeech benchmark among LLM-based ASR models and even outperforms the latest LLM-based audio-universal model trained on massive pair data. Finally, we explore the capability emergence of LLM-based ASR in the process of modal alignment. We hope that our study can facilitate the research on extending LLM with cross-modality capacity and shed light on the LLM-based ASR community.

LauraGPT: Listen, Attend, Understand, and Regenerate Audio with GPT

Generative Pre-trained Transformer (GPT) models have achieved remarkable performance on various natural language processing tasks. However, there has been limited research on applying similar frameworks to audio tasks. Previously proposed large language models for audio tasks either lack sufficient quantitative evaluations, or are limited to tasks for recognizing and understanding audio content, or significantly underperform existing state-of-the-art (SOTA) models. In this paper, we propose LauraGPT, a unified GPT model for audio recognition, understanding, and generation. LauraGPT is a versatile language model that can process both audio and text inputs and generate outputs in either modalities. It can perform a wide range of tasks related to content, semantics, paralinguistics, and audio-signal analysis. Some of its noteworthy tasks include automatic speech recognition, speech-to-text translation, text-to-speech synthesis, machine translation, speech enhancement, automated audio captioning, speech emotion recognition, and spoken language understanding. To achieve this goal, we use a combination of continuous and discrete features for audio. We encode input audio into continuous representations using an audio encoder and decode output audio from discrete codec codes. We then fine-tune a large decoder-only Transformer-based language model on multiple audio-to-text, text-to-audio, audio-to-audio, and text-to-text tasks using a supervised multitask learning approach. Extensive experiments show that LauraGPT achieves competitive or superior performance compared to existing SOTA models on various audio processing benchmarks.

SA-Paraformer: Non-autoregressive End-to-End Speaker-Attributed ASR

Joint modeling of multi-speaker ASR and speaker diarization has recently shown promising results in speaker-attributed automatic speech recognition (SA-ASR).Although being able to obtain state-of-the-art (SOTA) performance, most of the studies are based on an autoregressive (AR) decoder which generates tokens one-by-one and results in a large real-time factor (RTF). To speed up inference, we introduce a recently proposed non-autoregressive model Paraformer as an acoustic model in the SA-ASR model.Paraformer uses a single-step decoder to enable parallel generation, obtaining comparable performance to the SOTA AR transformer models. Besides, we propose a speaker-filling strategy to reduce speaker identification errors and adopt an inter-CTC strategy to enhance the encoder's ability in acoustic modeling. Experiments on the AliMeeting corpus show that our model outperforms the cascaded SA-ASR model by a 6.1% relative speaker-dependent character error rate (SD-CER) reduction on the test set. Moreover, our model achieves a comparable SD-CER of 34.8% with only 1/10 RTF compared with the SOTA joint AR SA-ASR model.

The second multi-channel multi-party meeting transcription challenge 2.0): A benchmark for speaker-attributed ASR

With the success of the first Multi-channel Multi-party Meeting Transcription challenge (M2MeT), the second M2MeT challenge (M2MeT 2.0) held in ASRU2023 particularly aims to tackle the complex task of speaker-attributed ASR (SA-ASR), which directly addresses the practical and challenging problem of "who spoke what at when" at typical meeting scenario. We particularly established two sub-tracks. 1) The fixed training condition sub-track, where the training data is constrained to predetermined datasets, but participants can use any open-source pre-trained model. 2) The open training condition sub-track, which allows for the use of all available data and models. In addition, we release a new 10-hour test set for challenge ranking. This paper provides an overview of the dataset, track settings, results, and analysis of submitted systems, as a benchmark to show the current state of speaker-attributed ASR.

FunCodec: A Fundamental, Reproducible and Integrable Open-source Toolkit for Neural Speech Codec

This paper presents FunCodec, a fundamental neural speech codec toolkit, which is an extension of the open-source speech processing toolkit FunASR. FunCodec provides reproducible training recipes and inference scripts for the latest neural speech codec models, such as SoundStream and Encodec. Thanks to the unified design with FunASR, FunCodec can be easily integrated into downstream tasks, such as speech recognition. Along with FunCodec, pre-trained models are also provided, which can be used for academic or generalized purposes. Based on the toolkit, we further propose the frequency-domain codec models, FreqCodec, which can achieve comparable speech quality with much lower computation and parameter complexity. Experimental results show that, under the same compression ratio, FunCodec can achieve better reconstruction quality compared with other toolkits and released models. We also demonstrate that the pre-trained models are suitable for downstream tasks, including automatic speech recognition and personalized text-to-speech synthesis. This toolkit is publicly available at https://github.com/alibaba-damo-academy/FunCodec.

CASA-ASR: Context-Aware Speaker-Attributed ASR

Recently, speaker-attributed automatic speech recognition (SA-ASR) has attracted a wide attention, which aims at answering the question ``who spoke what''. Different from modular systems, end-to-end (E2E) SA-ASR minimizes the speaker-dependent recognition errors directly and shows a promising applicability. In this paper, we propose a context-aware SA-ASR (CASA-ASR) model by enhancing the contextual modeling ability of E2E SA-ASR. Specifically, in CASA-ASR, a contextual text encoder is involved to aggregate the semantic information of the whole utterance, and a context-dependent scorer is employed to model the speaker discriminability by contrasting with speakers in the context. In addition, a two-pass decoding strategy is further proposed to fully leverage the contextual modeling ability resulting in a better recognition performance. Experimental results on AliMeeting corpus show that the proposed CASA-ASR model outperforms the original E2E SA-ASR system with a relative improvement of 11.76% in terms of speaker-dependent character error rate.