| import json, os, sys, tempfile, wave | |
| WAV_HEADER_SIZE = 44 | |
| WHISPER_CACHE = os.path.expanduser("~/.cache/whisper") | |
| def write_wav(path: str, raw_pcm: bytes, sample_rate: int = 16000): | |
| with wave.open(path, 'wb') as w: | |
| w.setnchannels(1) | |
| w.setsampwidth(2) | |
| w.setframerate(sample_rate) | |
| w.writeframes(raw_pcm) | |
| def transcribe(wav_path: str, model_size: str = "large-v3-turbo", language: str | None = None) -> dict: | |
| try: | |
| import whisper | |
| model = whisper.load_model(model_size, download_root=WHISPER_CACHE) | |
| opts = {} | |
| if language: | |
| opts["language"] = language | |
| result = model.transcribe(wav_path, **opts) | |
| return {"success": True, "text": result["text"].strip(), "language": result.get("language", "")} | |
| except ImportError: | |
| pass | |
| try: | |
| from faster_whisper import WhisperModel | |
| model_path = model_size | |
| if not os.path.isdir(model_path): | |
| local_path = os.path.expanduser(f"~/.cache/huggingface/hub/models--Systran--faster-whisper-{model_size}") | |
| if os.path.isdir(local_path): | |
| model_path = local_path | |
| model = WhisperModel(model_path, device="cpu", compute_type="int8") | |
| opts = {"beam_size": 1} | |
| if language: | |
| opts["language"] = language | |
| segments, info = model.transcribe(wav_path, **opts) | |
| text = " ".join(seg.text for seg in segments) | |
| return {"success": True, "text": text.strip(), "language": info.language} | |
| except ImportError: | |
| pass | |
| return {"success": False, "error": "Neither openai-whisper nor faster-whisper is installed. Run: pip install openai-whisper"} | |
| def main(): | |
| if len(sys.argv) < 2: | |
| print(json.dumps({"success": False, "error": "Usage: transcribe.py <wav_path> [--model <size>] [--language <code>]"})) | |
| sys.exit(1) | |
| wav_path = sys.argv[1] | |
| model_size = "large-v3-turbo" | |
| language = None | |
| i = 2 | |
| while i < len(sys.argv): | |
| if sys.argv[i] == "--model" and i + 1 < len(sys.argv): | |
| model_size = sys.argv[i + 1] | |
| i += 2 | |
| elif sys.argv[i] == "--language" and i + 1 < len(sys.argv): | |
| language = sys.argv[i + 1] | |
| i += 2 | |
| else: | |
| i += 1 | |
| if sys.argv[1] == "--stdin-pcm": | |
| sample_rate = int(sys.argv[2]) if len(sys.argv) > 2 else 16000 | |
| raw = sys.stdin.buffer.read() | |
| tmp = tempfile.mktemp(suffix=".wav") | |
| write_wav(tmp, raw, sample_rate) | |
| wav_path = tmp | |
| result = transcribe(wav_path, model_size, language) | |
| os.unlink(tmp) | |
| else: | |
| result = transcribe(wav_path, model_size, language) | |
| print(json.dumps(result)) | |
| if __name__ == "__main__": | |
| main() | |