# Pheme: Efficient and Conversational Speech Generation

arXiv: https://arxiv.org/abs/2401.02839 · HTML: https://mlsanity.com/paper/2401.02839/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 49 references that are in the map.
- First posted on arXiv 5 Jan 2024.

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech](https://mlsanity.com/paper/1904.02882/) (2019)
- [AudioLM: a Language Modeling Approach to Audio Generation](https://mlsanity.com/paper/2209.03143/) (2022)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [MLS: A Large-Scale Multilingual Dataset for Speech Research](https://mlsanity.com/paper/2012.03411/) (2020)
- [Libri-Light: A Benchmark for ASR with Limited or No Supervision](https://mlsanity.com/paper/1912.07875/) (2019)
- [MusicLM: Generating Music From Text](https://mlsanity.com/paper/2301.11325/) (2023)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [Scaling Speech Technology to 1,000+ Languages](https://mlsanity.com/paper/2305.13516/) (2023)
- [GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio](https://mlsanity.com/paper/2106.06909/) (2021)
- [BEATs: Audio Pre-Training with Acoustic Tokenizers](https://mlsanity.com/paper/2212.09058/) (2022)
- [AudioGen: Textually Guided Audio Generation](https://mlsanity.com/paper/2209.15352/) (2022)
- [AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head](https://mlsanity.com/paper/2304.12995/) (2023)
- [NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers](https://mlsanity.com/paper/2304.09116/) (2023)
- [Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal Supervision](https://mlsanity.com/paper/2302.03540/) (2023)
- [SoundStorm: Efficient Parallel Audio Generation](https://mlsanity.com/paper/2305.09636/) (2023)
- [Powerset multi-class cross entropy loss for neural speaker diarization](https://mlsanity.com/paper/2310.13025/) (2023)
- [InstructTTS: Modelling Expressive TTS in Discrete Latent Space with Natural Language Style Prompt](https://mlsanity.com/paper/2301.13662/) (2023)
- [Modular Deep Learning](https://mlsanity.com/paper/2302.11529/) (2023)

## Built on by (most cited first, 6 of 6)

- [SimpleSpeech: Towards Simple and Efficient Text-to-Speech with Scalar Latent Transformer Diffusion Models](https://mlsanity.com/paper/2406.02328/) (2024)
- [Generative Expressive Conversational Speech Synthesis](https://mlsanity.com/paper/2407.21491/) (2024)
- [TTSDS -- Text-to-Speech Distribution Score](https://mlsanity.com/paper/2407.12707/) (2024)
- [TTSDS2: Resources and Benchmark for Evaluating Human-Quality Text to Speech Systems](https://mlsanity.com/paper/2506.19441/) (2025)
- [Word-wise intonation model for cross-language TTS systems](https://mlsanity.com/paper/2409.20374/) (2024)
- [Efficient ASR Training with Conversations that Never Happened](https://mlsanity.com/paper/2606.03957/) (2026)

## Abstract (arXiv)

In recent years, speech generation has seen remarkable progress, now achieving one-shot generation capability that is often virtually indistinguishable from real human voice. Integrating such advancements in speech generation with large language models might revolutionize a wide range of applications. However, certain applications, such as assistive conversational systems, require natural and conversational speech generation tools that also operate efficiently in real time. Current state-of-the-art models like VALL-E and SoundStorm, powered by hierarchical neural audio codecs, require large neural components and extensive training data to work well. In contrast, MQTTS aims to build more compact conversational TTS models while capitalizing on smaller-scale real-life conversational speech data. However, its autoregressive nature yields high inference latency and thus limits its real-time usage. In order to mitigate the current limitations of the state-of-the-art TTS models while capitalizing on their strengths, in this work we introduce the Pheme model series that 1) offers compact yet high-performing models, 2) allows for parallel speech generation of 3) natural conversational speech, and 4) it can be trained efficiently on smaller-scale conversational data, cutting data demands by more than 10x but still matching the quality of the autoregressive TTS models. We also show that through simple teacher-student distillation we can meet significant improvements in voice quality for single-speaker setups on top of pretrained Pheme checkpoints, relying solely on synthetic speech generated by much larger teacher models. Audio samples and pretrained models are available online.