This is a text-to-speech model fine-tuned based on 300 hours of pure Arabic audio data, specifically designed to provide high-quality speech synthesis for Modern Standard Arabic with complete vowel symbols. It supports voice cloning and long text processing functions.

Audio ProcessingSafetensorsMultiple Languages

Zonos-v0.1 is a leading open-source text-to-speech (TTS) model, trained on over 200,000 hours of multilingual speech data, comparable in expressiveness and quality to top TTS providers. It supports zero-shot voice cloning, multilingual synthesis, and fine-grained audio control.

Audio ProcessingOnnx

Gödel Prover V2 is an open-source language model series that sets a new benchmark in automatic formal proof generation. It is built based on expert iteration and reinforcement learning processes, integrating three key innovations: scaffolded data synthesis, verifier-guided self-correction, and model averaging, which significantly improve the ability and efficiency of proving theorems.

Natural Language ProcessingTransformers

OpenAudio S1 is a leading text-to-speech (TTS) model trained on over 2 million hours of multilingual audio data. It supports 13 common languages, provides high-quality speech synthesis services, and supports rich emotional, intonation, and special effect markings.

This is a VITS model fine-tuned for Dhivehi speech synthesis, capable of converting Dhivehi text written in Tāna script into high-quality female voice audio. It is based on Meta's MMS-TTS architecture and trained using a carefully curated synthetic Dhivehi speech dataset.

Audio ProcessingTransformersOther

Text-to-speech model trained on 1000+ hours of Danish data, supporting natural conversation scenario speech synthesis

Audio ProcessingTransformers

xLAM-2 is a large action model built on an advanced data synthesis and training pipeline. It excels in multi-round dialogue and tool usage, and can transform user intentions into executable actions.

Natural Language ProcessingTransformersEnglish

The xLAM-2 series are Large Action Models (LAMs) built with advanced data synthesis and training pipelines, specializing in multi-turn conversations and tool usage, demonstrating exceptional performance in function calling and agent tasks.

Natural Language ProcessingTransformersEnglish

OASIS is a state-of-the-art code embedding model developed by Kwaipilot, integrating repository-level program analysis, the OASIS-instruct data synthesis algorithm, and a specialized fusion loss function, setting new benchmarks in code search efficiency and accuracy.

Natural Language ProcessingSafetensors

Experimental speech synthesis model based on Amis and Taroko languages, trained using the ithuan dataset

This is an Icelandic text-to-speech model fine-tuned based on facebook/mms-tts-isl, trained using the Talrómur dataset, specializing in female voice synthesis.

Audio ProcessingTransformersOther

NVIDIA NeMo Audio Codec 44khz is a neural audio codec that can compress audio into a quantized representation and can also be used as a vocoder for speech synthesis. This model is suitable for 44.1kHz full - bandwidth speech and supports speech data in 79 languages. Its performance may degrade on low - bandwidth speech or non - speech audio.

Audio ProcessingNemo

A Tamil speech synthesis model fine-tuned on the common_voice_17_0 dataset based on microsoft/speecht5_tts

Audio ProcessingTransformers

Uzbek speech synthesis model based on Microsoft SpeechT5, fine-tuned on an unknown dataset

Audio ProcessingTransformers

This model is a speech synthesis model fine-tuned on the Hindi validation set of the common_voice_17_0 dataset, based on microsoft/speecht5_tts.

Audio ProcessingTransformersOther

A Hindi speech synthesis model fine-tuned on the common_voice_17_0 dataset based on microsoft/speecht5_tts

Audio ProcessingTransformersOther

ReactionT5v2 is a pre - trained chemical retrosynthetic prediction model based on the Transformer architecture, specifically designed to predict the reactants of chemical reactions. This model is trained on the Open Reaction Database (ORD) and the USPTO_50k dataset, and can convert the SMILES of the target product into the possible SMILES of the reactants, providing support for the design of chemical synthesis routes.

Scientific ComputingTransformersEnglish

This is a male speech synthesis dataset specifically designed for Luxembourgish, as part of the Language Open Data initiative.

Audio ProcessingTransformersOther

BK-SDM-v2 is an efficient text-to-image synthesis model obtained by compressing SD-v2.1-base. It uses the block removal knowledge distillation technique to achieve effective distillation and retraining with limited data, providing a more lightweight solution for image generation.

Computer VisionDiffusers

XTTS is a text-to-speech model trained on a 40-hour dataset, supporting Russian and suitable for speech synthesis tasks in the legal domain.

Audio ProcessingTransformersOther