Batch Transcription

Source: 09_batch_transcription

Reads a local audio file, uploads it to the OpenAI Whisper API in one shot, and writes the transcript to a text file. Unlike the local speech transcription example, no model needs to be downloaded — the full audio is sent and a single transcript block is received back.

Running

melodium run 09_batch_transcription/Compo.toml \
  --input meeting.wav \
  --openai_key sk-...

[…] info: stt: reading audio file…
[…] info: stt: transcription complete

How it works

The pipeline is linear: read file → transcribe → write.

Block/Stream bridging

transcribe returns a Block<string> — a single value emitted once the full transcript is ready. Two downstream operations consume it, requiring two different adapters:

transcribe.transcript --> checkDone.value,check -> logDone.trigger
transcribe.transcript --> sttStream.block,stream -> write.text

• check<string>() discards the string value and emits Block<void>, used only to trigger the log
• stream<string>() converts the Block<string> into a Stream<string> that writeTextLocal can consume

The --> fan-out feeds both branches simultaneously from the single transcript block.

Dependencies

[dependencies]
std   = "0.10.1"  # core flows, logging, data structures
fs    = "0.10.1"  # local file I/O
ml    = "0.10.1"  # LLM, STT, TTS and local model inference