SuperAPI AI Directory
Partner with us中文AI newsFavorites

AI audio

1,529 results · Page 15 of 43

YuE2

AI audio

YuE2 is an open-source music generation model from Multimodal Art Projection (M-A-P) that turns lyrics and style prompts into complete songs containing vocals and accompaniment. The public model checkpoint is YuE2-3B; official inference first writes the melody and chords (ABC notation plan) and then renders the audio, supporting covers, reharmonization, and agentic editing. This site is its online chat studio: paste lyrics and style prompts, and generate and repeatedly revise songs through conversation. Released demos cover English, Mandarin, Japanese, and Spanish. On the WildSongBench (September 12, 2026) standard split, the SongBench average score is 6.7316, with the best-of-eight at 6.9632, higher than the reported Suno v5 (6.8721) and Suno v6 (6.5562). The weights are hosted on Hugging Face, and local running requires Linux, Python 3.10+, and a 24 GB NVIDIA GPU (BF16). The site offers paid plans involving credits, downloads, and commercial licensing, etc.; the live plan table is authoritative.

Read more

Официальный сайт

AI audio

Music Labs is an online AI music creation platform where users can generate songs with vocals, accompaniment, and lyrics through neural networks, supporting Russian and more than 50 other languages. The platform also offers covers, voice cloning (singing with your own recorded voice), AI voiceover (341 voice timbres), sound effect generation, stem extraction and mixing, a music editor, and cover art generation. Registration comes with 10 free credits, which can be used for about 6 songs. It supports payment via Russian bank cards or SBP, can be accessed without a VPN, and downloads are available in MP3 and WAV formats.

Read more

文字MP3專業版

AI audio

Text-to-MP3 Professional Edition is AI dubbing software launched by Internet Intelligence, with the main focus of "dubbing through text editing." It operates like using Word, and dubbing files can be produced without editing backgrounds. Users only need to enter text and insert sound effect tags (such as voice actor, speech rate, volume, background sound) to complete multilingual, multi-person dialogue dubbing content. It offers two sound quality levels, Gold and Advanced: Gold sound quality supports 100 languages and more than 600 voice actors, and requires an internet connection; Advanced sound quality supports 2 languages, Chinese and English, with a total of 2 voice actors, and can be used offline. It supports mixed reading of Chinese and English as well as dates, numbers, and more, allows adjustment of speech rate, pitch, pause length, and the tone of some voice actors, and can insert sound effects and background sounds. It includes 12 sample templates, covering short stories, children's teaching, telephone switchboard, building announcements, and more. Output formats are MP3 or WAV, and it supports Windows 10 and Windows 11. Customized AI voice actor model training and dubbing services are also provided.

Read more

Ahsk

AI audio

Ahsk is an AI voice dictation app for Mac, Windows, and iPhone. Hold down a shortcut key (fn on Mac, Ctrl + Win on Windows) and speak; when you release it, the cleaned-up text appears at the current cursor position, with support for over 50 languages and translation while speaking. Speech is transcribed in the cloud by Whisper large-v3, then punctuation is automatically cleaned up, paragraphs are split, filler words are removed, and the text is polished according to the selected tone mode (such as Informal, Formal, Mail). It supports continued recording after closing the lid or switching headphones, recognition of accented English, and a custom dictionary. It can also generate meeting notes from Zoom, Teams, and Meet, and on Mac it offers the Alan (beta) voice assistant. Start for free, Pro is €9.99 per month. It requires an internet connection, has no offline mode, and has no Android version.

Read more