Moonshine ASR models running on-device, in your web browser.
Generate audio from text or file
Transcribe spoken Russian into text
Convert text into speech in Japanese
Transcribe audio or YouTube videos into text
Identify speakers in an audio file
Kokoro is an open-weight TTS model with 82 million parameters.
Generate audio from text with customizable voice
A demo of Indic Parler-TTS
Generate Vietnamese speech from text and reference audio
Transcribe Persian audio files into text
Pyxilab's Pyx r1-voice demo
"Designed for all users, including those with disabilities."
Moonshine Web is a web-based tool designed for speech synthesis, enabling users to transcribe speech to text using audio files. It leverages Moonshine ASR models, which run on-device directly in your web browser, ensuring privacy and efficiency. This tool is ideal for converting audio content into readable text quickly and accurately.
What audio formats does Moonshine Web support?
Moonshine Web supports a variety of audio formats, including WAV, MP3, and AAC. For a complete list, refer to the official documentation.
Can I customize the transcription settings?
Yes, Moonshine Web allows users to customize settings such as speech recognition models and noise reduction to improve transcription accuracy.
Is my data kept private since it runs on-device?
Absolutely. Since Moonshine Web processes audio files locally in your browser, your data never leaves your device, ensuring complete privacy.