Transcribe audio files into text using the z-ai-web-dev-sdk.
npx skills add https://github.com/skkarki/Sanam --skill asr-skkarki
Please help me install this Agent Skill. Skill: ASR Source: https://github.com/skkarki/Sanam/tree/main/skills/ASR Command: npx skills add https://github.com/skkarki/Sanam --skill asr-skkarki
This Skill addresses the need to convert spoken audio into accurate, searchable text for transcripts, captions, accessibility, and analytics.
Transcribe a file: z-ai asr --file ./audio.wav Transcribe from base64: z-ai asr --base64 "BASE64_ENCODED_AUDIO" -o transcript.json Stream transcription: z-ai asr -f ./audio.wav --stream
High-intent search queries and answers about installing and using this skill.
To transcribe audio files to text using a backend SDK, use the z-ai-web-dev-sdk to process formats like WAV, MP3, and M4A. It handles base64 encoding and provides error handling for batch processing and real-time transcription.
Yes, you can transcribe multiple audio files from a directory using the batch processing feature. The CLI supports directory-wide tasks, allowing you to automate transcription for common formats like WAV and MP3.
Yes, the speech-to-text CLI supports real-time streaming transcription. You can use the stream flag with the CLI to process audio files like WAV and output text dynamically as the audio is processed.
The supported audio formats for transcription and accessibility are WAV, MP3, and M4A. These common formats are processed into clean text for transcripts, captions, and analytics via the z-ai-web-dev-sdk.
You need base64 encoding to process audio for speech-to-text conversion when passing audio data directly through the CLI. The tool accepts a base64 encoded string as an alternative to providing a file path.
The best way to automate meeting notes transcription from voice memos is using the CLI for batch jobs. It transcribes voice memos into searchable text, transforming interviews and podcasts into text for analysis.
felipesantiago-coderConvert base64 encoded audio files into written text transcriptions.
ahostbrExtract YouTube subtitles into timestamped markdown transcripts via yt-dlp.
nidhalboumaiza-0Transcribe base64-encoded audio into text using the z-ai-web-dev-sdk.
elevenlabsTranscribe audio to text using ElevenLabs Scribe v2.
