vikarti.anatra/AudioSumma
1
0
Fork 0
mirror of https://github.com/patw/AudioSumma.git synced 2025-09-01 10:09:56 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Skip transcribe step if it's already done. Show each part being summarized.

Browse source
This commit is contained in:
Pat Wendorf 2025-02-03 16:30:52 -05:00
parent 2d329d4806
commit 933b9e24d4
1 changed files with 9 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
summarize.py
View file

@ -57,6 +57,14 @@ def process_wav_files():
"""Process WAV files: trim silence and transcribe.""" """Process WAV files: trim silence and transcribe."""
wav_files = [f for f in os.listdir(".") if f.endswith(".wav")] wav_files = [f for f in os.listdir(".") if f.endswith(".wav")]
for wav_file in wav_files: for wav_file in wav_files:
# Generate the expected transcript filename
transcript_file = os.path.splitext(wav_file)[0] + ".tns"
# Check if transcript already exists
if os.path.exists(transcript_file):
print(f"Transcript already exists for {wav_file}, skipping transcription")
continue
print("Trimming silence: " + wav_file) print("Trimming silence: " + wav_file)
trim_silence(wav_file) trim_silence(wav_file)
@ -95,6 +103,7 @@ def summarize_transcripts():
with open(summary_filename, "a") as md_file: with open(summary_filename, "a") as md_file:
for i, chunk in enumerate(chunked_data): for i, chunk in enumerate(chunked_data):
print("Processing part " + str(i))
summary = llm_local(SUMMARY_PROMPT.format(chunk=chunk)) summary = llm_local(SUMMARY_PROMPT.format(chunk=chunk))
facts = llm_local(FACT_PROMPT.format(chunk=chunk)) facts = llm_local(FACT_PROMPT.format(chunk=chunk))
sentiment = llm_local(SENTIMENT_PROMPT.format(chunk=chunk)) sentiment = llm_local(SENTIMENT_PROMPT.format(chunk=chunk))