17 个工具
开源 AI 音频、语音与音乐工具
语音识别、语音合成、声音克隆、人声分离和音乐生成。
WhisperOpenAI’s open-source speech recognition and translation model.MITwhisper.cppWhisper in plain C/C++ — fast on CPUs, Apple Silicon and mobile.MITfaster-whisperWhisper reimplemented on CTranslate2 — up to 4× faster with less memory.MITWhisperXFast transcription with word-level timestamps and speaker diarization.BSD-2-ClauseCoqui TTSDeep-learning text-to-speech toolkit, including the XTTS voice cloning model.MPL-2.0GPT-SoVITSFew-shot voice cloning and TTS with a full training WebUI.MITRVC WebUIRetrieval-based voice conversion — train a voice model and convert singing or speech.MITApplioA modern, easier RVC-based voice conversion app.MITF5-TTSZero-shot voice cloning TTS based on flow matching.MITFish SpeechMultilingual TTS and voice cloning from Fish Audio.Apache-2.0KokoroA tiny 82M-parameter TTS model with surprisingly natural voices.Apache-2.0ChatTTSConversational TTS for dialogue, with laughter and pauses.AGPL-3.0BarkSuno’s generative audio model: speech, music snippets and sound effects.MITAudioCraft (MusicGen)Meta’s library for MusicGen music generation and AudioGen sound effects.MITStable Audio ToolsTraining and inference code for Stable Audio Open.MITDemucsSeparate vocals, drums, bass and other stems from any song.MITUltimate Vocal RemoverDesktop GUI for vocal removal and stem separation.MIT