Transcribe and translate spoken audio into text with timestamps.
npx skills add https://github.com/MarbleSodas/Mavis --skill whisper-marblesodas
Please help me install this Agent Skill. Skill: whisper Source: https://github.com/MarbleSodas/Mavis/tree/main/optional-skills/mlops/whisper Command: npx skills add https://github.com/MarbleSodas/Mavis --skill whisper-marblesodas
Whisper solves the problem of turning spoken audio into readable text, including multilingual transcription and translation to English, so you can document meetings, podcasts, or media without manual typing.
Example use case: Convert a recorded multilingual team meeting into structured notes by transcribing the audio and translating the result to English for faster review.
Use the whisper skill to transcribe and translate your audio file to English, and return the generated text.
High-intent search queries and answers about installing and using this skill.
To transcribe and translate audio files into English text, use speech-to-text processing to either faithfully transcribe the spoken language or translate non-English audio directly into English written text.
Yes, automatic language identification is supported during audio transcription, enabling the processing of multilingual content across 99 languages without requiring prior language specification.
Generate timestamps for podcast and meeting transcriptions by configuring decoding options to produce word-level or segment-level timestamps alongside the transcribed text output.
Speech-to-text transcription supports technical content refinement by accepting an initial prompt, which guides the decoding process to improve accuracy for specialized terminology and specific contexts.
The best way to transcribe multilingual team meetings into structured notes is to apply speech-to-text transcription with translation to English, enabling faster review and documentation of the audio content.
kaivyyTranscribe audio files with OpenAI Whisper and synthesize speech from text.
RobbyMo81Transcribe local audio files into text and subtitle formats.
davpatel605-beepTranscribe spoken audio into timestamped text across 99 languages.
aztr0nutzsTranscribe live audio streams and pre-recorded files via Deepgram SDK.
