diff --git a/scripts/recording_audio.py b/scripts/recording_audio.py index 1481273fc..245f1babd 100644 --- a/scripts/recording_audio.py +++ b/scripts/recording_audio.py @@ -4,12 +4,13 @@ Generate TTS audio files for recording commentary using OpenAI's API. Usage: python scripts/recording_audio.py path/to/recording.md """ +import argparse import os import re import sys -import argparse -import requests from pathlib import Path + +import requests from dotenv import load_dotenv # Load environment variables from .env file @@ -20,59 +21,55 @@ OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY") OUTPUT_DIR = "aider/website/assets/audio" VOICE = "onyx" # Options: alloy, echo, fable, onyx, nova, shimmer + def extract_recording_id(markdown_file): """Extract recording ID from the markdown file path.""" return Path(markdown_file).stem + def extract_commentary(markdown_file): """Extract commentary markers from markdown file.""" - with open(markdown_file, 'r') as f: + with open(markdown_file, "r") as f: content = f.read() - + # Find Commentary section - commentary_match = re.search(r'## Commentary\s+(.*?)(?=##|\Z)', content, re.DOTALL) + commentary_match = re.search(r"## Commentary\s+(.*?)(?=##|\Z)", content, re.DOTALL) if not commentary_match: print(f"No Commentary section found in {markdown_file}") return [] - + commentary = commentary_match.group(1).strip() - + # Extract timestamp-message pairs markers = [] - for line in commentary.split('\n'): + for line in commentary.split("\n"): line = line.strip() - if line.startswith('- '): + if line.startswith("- "): line = line[2:] # Remove the list marker - match = re.match(r'(\d+):(\d+)\s+(.*)', line) + match = re.match(r"(\d+):(\d+)\s+(.*)", line) if match: minutes, seconds, message = match.groups() time_in_seconds = int(minutes) * 60 + int(seconds) markers.append((time_in_seconds, message)) - + return markers + def generate_audio_openai(text, output_file): """Generate audio using OpenAI TTS API.""" if not OPENAI_API_KEY: print("Error: OPENAI_API_KEY environment variable not set") return False - + url = "https://api.openai.com/v1/audio/speech" - headers = { - "Authorization": f"Bearer {OPENAI_API_KEY}", - "Content-Type": "application/json" - } - data = { - "model": "tts-1", - "input": text, - "voice": VOICE - } - + headers = {"Authorization": f"Bearer {OPENAI_API_KEY}", "Content-Type": "application/json"} + data = {"model": "tts-1", "input": text, "voice": VOICE} + try: response = requests.post(url, headers=headers, json=data) - + if response.status_code == 200: - with open(output_file, 'wb') as f: + with open(output_file, "wb") as f: f.write(response.content) return True else: @@ -82,36 +79,41 @@ def generate_audio_openai(text, output_file): print(f"Exception during API call: {e}") return False + def main(): - parser = argparse.ArgumentParser(description='Generate TTS audio for recording commentary.') - parser.add_argument('markdown_file', help='Path to the recording markdown file') - parser.add_argument('--voice', default=VOICE, help=f'OpenAI voice to use (default: {VOICE})') - parser.add_argument('--output-dir', default=OUTPUT_DIR, help=f'Output directory (default: {OUTPUT_DIR})') - parser.add_argument('--dry-run', action='store_true', help='Print what would be done without generating audio') - + parser = argparse.ArgumentParser(description="Generate TTS audio for recording commentary.") + parser.add_argument("markdown_file", help="Path to the recording markdown file") + parser.add_argument("--voice", default=VOICE, help=f"OpenAI voice to use (default: {VOICE})") + parser.add_argument( + "--output-dir", default=OUTPUT_DIR, help=f"Output directory (default: {OUTPUT_DIR})" + ) + parser.add_argument( + "--dry-run", action="store_true", help="Print what would be done without generating audio" + ) + args = parser.parse_args() - + # Update globals with any command line overrides global VOICE VOICE = args.voice - + recording_id = extract_recording_id(args.markdown_file) print(f"Processing recording: {recording_id}") - + # Create output directory output_dir = os.path.join(args.output_dir, recording_id) if not args.dry_run: os.makedirs(output_dir, exist_ok=True) - + # Extract commentary markers markers = extract_commentary(args.markdown_file) - + if not markers: print("No commentary markers found!") return - + print(f"Found {len(markers)} commentary markers") - + # Generate audio for each marker for time_sec, message in markers: minutes = time_sec // 60 @@ -119,7 +121,7 @@ def main(): timestamp = f"{minutes:02d}-{seconds:02d}" filename = f"{timestamp}.mp3" output_file = os.path.join(output_dir, filename) - + print(f"Marker at {minutes}:{seconds:02d} - {message}") if args.dry_run: print(f" Would generate: {output_file}") @@ -131,5 +133,6 @@ def main(): else: print(f" ✗ Failed to generate audio") + if __name__ == "__main__": main()