SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,738 results · Page 651 of 993

Fine Voice

AI audio

Fine Voice is an online AI voice generator that requires no installation and can perform text-to-speech, instant voice cloning and voice design directly in the browser. It provides realistic speech with emotion, pacing and emphasis, supports multiple languages and accents, and can be used for voiceovers in content such as videos, podcasts, apps and courses. Free accounts have a limit of 120 characters per conversion, while paid plans and credit packs can raise this to 1,000 characters per conversion; exported audio comes with commercial usage rights, and a low-latency API is offered to developers so speech can be streamed into products.

Read more

FinGPT

AI finance

FinGPT is an open-source financial large language model project developed and maintained by the AI4Finance Foundation. The project aims to provide the financial sector with an open-source large language model solution that can be quickly fine-tuned at a lower cost, addressing the issues of highly dynamic financial data and the high training costs of traditional models. FinGPT emphasizes lightweight adaptation, supports fine-tuning at a lower cost using methods such as LoRA, and provides capabilities including financial sentiment analysis, financial forecasting, and instruction fine-tuning benchmark evaluation. The project has released trained models, datasets, and demo spaces on HuggingFace, where users can try FinGPT-Forecaster for stock price trend prediction. FinGPT adopts a five-layer full-stack framework, covering the data source, data engineering, large language model, task, and application layers.

Read more

Fireflies APP最新版免费下载-安卓和iOS手机版

AI apps

Fireflies is a powerful AI meeting note-taking and analysis tool. It can automatically record and transcribe meeting content from a variety of video conferencing applications (such as Zoom, Google Meet, etc.), generate text transcripts, and quickly extract key information from meetings, such as action items and tasks. It supports team collaboration, allowing users to create soundbites and share important moments from meetings, and can send notes to collaboration applications such as Slack and Notion.

Read more

FireRedTTS2

AI audio

FireRedTTS2 is an open-source PyTorch long-conversation speech synthesis system from FireRedTeam, aimed at podcast and chatbot scenarios, supporting multi-speaker dialogue generation. It uses a 12.5Hz streaming speech tokenizer and a dual Transformer architecture, operating on text-speech interleaved sequences, and can generate sentence by sentence. The official claim is that first-packet latency is as low as 140ms on an L20 GPU. The system supports multiple languages including English, Chinese, Japanese, Korean, French, German, and Russian, and supports zero-shot voice cloning for cross-lingual and code-switching scenarios. It currently supports 3-minute dialogues with 4 speakers, and can be extended to longer dialogues with more speakers by expanding the training corpus. The project provides pretrained checkpoints, inference code, Web UI tools, fine-tuning code, and tutorials, and supports bf16 inference to reduce VRAM usage.

Read more