datasci/verbanote-server
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Add info logging to failable functions

Browse source
This commit is contained in:
Marty Oehme 2023-08-23 15:10:41 +02:00
parent 6cccf00df4
commit f7b0d48c6f
Signed by: Marty
GPG key ID: EDBF2ED917B2EF6A
2 changed files with 7 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
verbanote/file_operations.py
View file

@ -23,6 +23,7 @@ def upload_to_oxo(file: Path, url: str = "https://0x0.st", expires: int = 2) ->
) )
if not resp.ok: if not resp.ok:
raise requests.exceptions.HTTPError() raise requests.exceptions.HTTPError()
logging.info(f"Uploaded file {file} to {str(resp.content)}")
return str(resp.content) return str(resp.content)

6
verbanote/process.py
View file

@ -1,3 +1,4 @@
import logging
import os import os
import re import re
import json import json
@ -43,11 +44,13 @@ def transcribe(
f = {Path.joinpath(output_path, str(i))} f = {Path.joinpath(output_path, str(i))}
audio_f = f"{f}.wav" audio_f = f"{f}.wav"
json_f = f"{f}.json" json_f = f"{f}.json"
logging.info(f"Starting transcription of {str(audio_f)}...")
result = model.transcribe( result = model.transcribe(
audio=audio_f, language=lang, word_timestamps=word_timestamps audio=audio_f, language=lang, word_timestamps=word_timestamps
) )
with open(json_f, "w") as outfile: with open(json_f, "w") as outfile:
json.dump(result, outfile, indent=4) json.dump(result, outfile, indent=4)
logging.info(f"Transcription written to {str(json_f)}.")
# TODO clean up this mess # TODO clean up this mess
@ -65,6 +68,7 @@ def output_txt(diarized_groups: list, transcription_path: Path) -> TxtTranscript
with open(f"{Path.joinpath(transcription_path, str(gidx))}.json") as f: with open(f"{Path.joinpath(transcription_path, str(gidx))}.json") as f:
captions = json.load(f)["segments"] captions = json.load(f)["segments"]
logging.info(f"Loaded {fname} for transcription...")
if captions: if captions:
speaker = g[0].split()[-1] speaker = g[0].split()[-1]
@ -95,6 +99,7 @@ def _add_audio_silence(audiofile) -> Path:
audio = spacer.append(audio, crossfade=0) audio = spacer.append(audio, crossfade=0)
out_file = Path.joinpath(Path(os.path.dirname(audiofile)), "interview_prepend.wav") out_file = Path.joinpath(Path(os.path.dirname(audiofile)), "interview_prepend.wav")
audio.export(out_file, format="wav") audio.export(out_file, format="wav")
logging.info(f"Exported audiofile with silence prepended to {fname}.")
return out_file return out_file
@ -113,6 +118,7 @@ def _save_individual_audio_files(
audio[start:end].export( audio[start:end].export(
f"{Path.joinpath(output_path, str(gidx))}.wav", format="wav" f"{Path.joinpath(output_path, str(gidx))}.wav", format="wav"
) )
logging.info(f"Exported audiopart {gidx} of {len(groups)} to {fname}.")
def _group_speakers(diarization_file: Path) -> list: def _group_speakers(diarization_file: Path) -> list: