Transcribe, analyze, and generate speech from audio via the Google Gemini API.
npx skills add https://github.com/alex-tgk/saasaas --skill gemini-audio
Please help me install this Agent Skill. Skill: gemini-audio Source: https://github.com/alex-tgk/saasaas/tree/main/.claude/skills/gemini-audio Command: npx skills add https://github.com/alex-tgk/saasaas --skill gemini-audio
Gemini Audio provides transcription, analysis, and summarization of audio, plus text-to-speech generation. It streamlines workflows for podcasts, interviews, meetings, and multimedia content by turning audio into searchable text and actionable insights.
Configure GEMINI_API_KEY, then run transcribe.py or generate-speech.py to process audio files or synthesize speech.
High-intent search queries and answers about installing and using this skill.
Transcription with timestamps and multi-speaker support is a core feature of Gemini Audio. Configure your GEMINI_API_KEY, use the transcribe.py script with WAV, MP3, AAC, FLAC, OGG, or AIFF files (up to 9.5 hours), and the API returns text with timing markers and speaker labels for interviews, meetings, and podcasts.
Gemini Audio supports WAV, MP3, AAC, FLAC, OGG, and AIFF formats. Files can be up to 9.5 hours long and processed via file-based or inline data inputs through the Google Gemini API.
Yes. The TTS generation feature in Gemini Audio allows you to control voice style, pace, tone, and accent when synthesizing speech from text. Run generate-speech.py to produce audio with your specified parameters.
Gemini Audio analyzes audio content to produce summaries and key-point extraction alongside transcription. Process podcasts, interviews, or meetings with the Gemini API to get both searchable text and actionable insights in a single workflow.
Yes. Gemini Audio supports multilingual transcription, so you can process audio in different languages and receive transcripts in the source language or translated output.
Yes. Beyond transcription, Gemini Audio identifies and analyzes non-speech audio such as music, ambient sounds, and background noise, providing detailed analysis alongside speaker and content transcription.
duotifyTranscribe audio files into Traditional Chinese Markdown transcripts with speaker labels.
vreExtract YouTube video metadata and transcripts into structured Markdown documents.
steipeteTranscribe audio files into text using the OpenAI Whisper API.
jayll1303Perform offline speech processing with ONNX models via sherpa-onnx.
