kunthawat/ALwrity
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Alwrity - Bug fixes

Browse Source
This commit is contained in:
AjaySi
2024-04-12 17:36:37 +05:30
parent cf6516eeee
commit e3c3c03729
18 changed files with 81 additions and 81 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
lib/gpt_providers/audio_to_text_generation/stt_audio_blog.py
View File

@@ -79,7 +79,7 @@ def speech_to_text(video_url, output_path='.'):
logger.info("Transcribing using OpenAI's Whisper model.")
transcript = client.audio.transcriptions.create(
model="whisper-1",
file=open(audio_file, "rb"),
file=open(audio_file, "rb", encoding="utf-8"),
response_format="text"
)
logger.info(f"\nYouTube video transcription:\n{yt.title}\n{transcript}\n")
@@ -133,7 +133,7 @@ def long_video(temp_file_name):
for i, chunk in enumerate(chunks):
with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as audio_chunk_file:
chunk.write_audiofile(audio_chunk_file.name, codec="mp3")
with open(audio_chunk_file.name, "rb") as audio_file:
with open(audio_chunk_file.name, "rb", encoding="utf-8") as audio_file:
# Transcribe each chunk using OpenAI's Whisper API
app.logger.info(f"Transcribing chunk {i+1}/{len(chunks)}")
transcript = openai.Audio.transcribe("whisper-1", audio_file)