LLM-to-Speech: A Synthetic Data Pipeline for Training Dialectal Text-to-Speech Models

• URL: http://arxiv.org/abs/2602.15675v1
• Date: Tue, 17 Feb 2026 15:58:27 GMT
• Title: LLM-to-Speech: A Synthetic Data Pipeline for Training Dialectal Text-to-Speech Models
• Authors: Ahmed Khaled Khamis, Hesham Ali,
• Abstract summary: We introduce NileTTS: 38 hours of transcribed speech from two speakers across diverse domains.<n>Large language models (LLM) generate Egyptian Arabic content, which is converted to natural speech using audio synthesis tools.<n>We fine-tune XTTS v2, a state-of-the-art multilingual TTS model, on our dataset and evaluate against the baseline model trained on other Arabic dialects.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite the advances in neural text to speech (TTS), many Arabic dialectal varieties remain marginally addressed, with most resources concentrated on Modern Spoken Arabic (MSA) and Gulf dialects, leaving Egyptian Arabic -- the most widely understood Arabic dialect -- severely under-resourced. We address this gap by introducing NileTTS: 38 hours of transcribed speech from two speakers across diverse domains including medical, sales, and general conversations. We construct this dataset using a novel synthetic pipeline: large language models (LLM) generate Egyptian Arabic content, which is then converted to natural speech using audio synthesis tools, followed by automatic transcription and speaker diarization with manual quality verification. We fine-tune XTTS v2, a state-of-the-art multilingual TTS model, on our dataset and evaluate against the baseline model trained on other Arabic dialects. Our contributions include: (1) the first publicly available Egyptian Arabic TTS dataset, (2) a reproducible synthetic data generation pipeline for dialectal TTS, and (3) an open-source fine-tuned model. All resources are released to advance Egyptian Arabic speech synthesis research.

Related papers

• ArFake: A Multi-Dialect Benchmark and Baselines for Arabic Spoof-Speech Detection [2.5962590697722447]
  We introduce the first multi-dialect Arabic spoofed speech dataset.<n>Our results demonstrate that FishSpeech outperforms other TTS models in Arabic voice cloning on the Casablanca corpus.
  arXiv  Detail & Related papers  (2025-09-26T18:11:20Z)
• OpusLM: A Family of Open Unified Speech Language Models [56.14140121061921]
  The OpusLMs are continuously pre-trained on 213K hours of speech-text pairs and 292B text-only tokens.<n>Technically, this paper articulates our SpeechLM designs on tokenization, multi-stream language models, and multi-stage training strategies.
  arXiv  Detail & Related papers  (2025-06-21T06:30:59Z)
• CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models [74.80386066714229]
  We present an improved streaming speech synthesis model, CosyVoice 2.<n>Specifically, we introduce finite-scalar quantization to improve codebook utilization of speech tokens.<n>We develop a chunk-aware causal flow matching model to support various synthesis scenarios.
  arXiv  Detail & Related papers  (2024-12-13T12:59:39Z)
• VoiceTextBlender: Augmenting Large Language Models with Speech Capabilities via Single-Stage Joint Speech-Text Supervised Fine-Tuning [64.56272011710735]
  We propose a novel single-stage joint speech-text SFT approach on the low-rank adaptation (LoRA) of the large language models (LLMs) backbone.<n>Compared to previous SpeechLMs with 7B or 13B parameters, our 3B model demonstrates superior performance across various speech benchmarks.
  arXiv  Detail & Related papers  (2024-10-23T00:36:06Z)
• Towards Zero-Shot Text-To-Speech for Arabic Dialects [16.10882912169842]
  Zero-shot multi-speaker text-to-speech (ZS-TTS) systems have advanced for English, however, it still lags behind due to insufficient resources. We address this gap for Arabic by first adapting an existing dataset to suit the needs of speech synthesis. We employ a set of Arabic dialect identification models to explore the impact of pre-defined dialect labels on improving the ZS-TTS model in a multi-dialect setting.
  arXiv  Detail & Related papers  (2024-06-24T15:58:15Z)
• CoSTA: Code-Switched Speech Translation using Aligned Speech-Text Interleaving [61.73180469072787]
  We focus on the problem of spoken translation (ST) of code-switched speech in Indian languages to English text. We present a new end-to-end model architecture COSTA that scaffolds on pretrained automatic speech recognition (ASR) and machine translation (MT) modules. COSTA significantly outperforms many competitive cascaded and end-to-end multimodal baselines by up to 3.5 BLEU points.
  arXiv  Detail & Related papers  (2024-06-16T16:10:51Z)
• ArTST: Arabic Text and Speech Transformer [2.53638770809417]
  We present ArTST, a pre-trained Arabic text and speech transformer. It supports open-source speech technologies for the Arabic language.
  arXiv  Detail & Related papers  (2023-10-25T13:20:54Z)
• AudioPaLM: A Large Language Model That Can Speak and Listen [79.44757696533709]
  We introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models. It can process and generate text and speech with applications including speech recognition and speech-to-speech translation.
  arXiv  Detail & Related papers  (2023-06-22T14:37:54Z)
• Textless Speech-to-Speech Translation With Limited Parallel Data [51.3588490789084]
  PFB is a framework for training textless S2ST models that require just dozens of hours of parallel speech data. We train and evaluate our models for English-to-German, German-to-English and Marathi-to-English translation on three different domains.
  arXiv  Detail & Related papers  (2023-05-24T17:59:05Z)
• ClArTTS: An Open-Source Classical Arabic Text-to-Speech Corpus [3.1925030748447747]
  We present a speech corpus for Classical Arabic Text-to-Speech (ClArTTS) to support the development of end-to-end TTS systems for Arabic. The speech is extracted from a LibriVox audiobook, which is then processed, segmented, and manually transcribed and annotated. The final ClArTTS corpus contains about 12 hours of speech from a single male speaker sampled at 40100 kHz.
  arXiv  Detail & Related papers  (2023-02-28T20:18:59Z)
• End-to-End Speech Translation of Arabic to English Broadcast News [2.375764121997739]
  Speech translation (ST) is the task of translating acoustic speech signals in a source language into text in a foreign language. This paper presents our efforts towards the development of the first Broadcast News end-to-end Arabic to English speech translation system.
  arXiv  Detail & Related papers  (2022-12-11T11:35:46Z)
• Textless Speech-to-Speech Translation on Real Data [49.134208897722246]
  We present a textless speech-to-speech translation (S2ST) system that can translate speech from one language into another language. We tackle the challenge in modeling multi-speaker target speech and train the systems with real-world S2ST data.
  arXiv  Detail & Related papers  (2021-12-15T18:56:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.