How do I extract text from audio?
Transcribe the audio with a Whisper ASR capability. a2a.computer plans this as a single governed hop and returns an evidence receipt for the transcription.
from audioto text
Available paths (live from the planner)
#11 hoploss 10%~30srisk medium
audio→transcribe_audio→text
How a2a.computer knows this
A single-hop governed capability (huggingface.transcribe_audio). The transcript itself is never written to evidence — only the model, language, segment count, and latency are recorded.
For agents
Capabilities:
chp.adapters.huggingface.transcribe_audioMCP tools:
find_transmutation_pathsestimate_transmutationrealize_stateretrieve_evidenceLimitations & uncertainty
- –ASR accuracy varies with audio quality, accent, and domain vocabulary.
- –The transcript is returned to the caller but deliberately excluded from the evidence record.