lyrics-transcriber 0.30.0__py3-none-any.whl → 0.32.1__py3-none-any.whl

lyrics_transcriber/transcribers/audioshake.py

@@ -1,151 +1,217 @@
+from dataclasses import dataclass
 import requests
 import time
 import os
-import json
-from .base import BaseTranscriber
+from typing import Dict, Optional, Any, Union
+from pathlib import Path
+from lyrics_transcriber.types import TranscriptionData, LyricsSegment, Word
+from lyrics_transcriber.transcribers.base_transcriber import BaseTranscriber, TranscriptionError
 
 
-class AudioShakeTranscriber(BaseTranscriber):
-    """Transcription service using AudioShake's API."""
-
-    def __init__(self, api_token=None, logger=None, output_prefix=None):
-        super().__init__(logger)
-        self.api_token = api_token or os.getenv("AUDIOSHAKE_API_TOKEN")
-        self.base_url = "https://groovy.audioshake.ai"
-        self.output_prefix = output_prefix
-
-        if not self.api_token:
-            raise ValueError("AudioShake API token must be provided either directly or via AUDIOSHAKE_API_TOKEN env var")
-
-    def get_name(self) -> str:
-        return "AudioShake"
-
-    def transcribe(self, audio_filepath: str) -> dict:
-        """
-        Transcribe an audio file using AudioShake API.
-
-        Args:
-            audio_filepath: Path to the audio file to transcribe
-
-        Returns:
-            Dict containing:
-                - segments: List of segments with start/end times and word-level data
-                - text: Full text transcription
-                - metadata: Dict of additional info
-        """
-        self.logger.info(f"Starting transcription for {audio_filepath} using AudioShake API")
-
-        # Start job and get results
-        job_id = self.start_transcription(audio_filepath)
-        result = self.get_transcription_result(job_id)
-
-        # Add metadata to the result
-        result["metadata"] = {
-            "service": self.get_name(),
-            "language": "en",  # AudioShake currently only supports English
-        }
-
-        return result
+@dataclass
+class AudioShakeConfig:
+    """Configuration for AudioShake transcription service."""
 
-    def start_transcription(self, audio_filepath: str) -> str:
-        """Starts the transcription job and returns the job ID."""
-        # Step 1: Upload the audio file
-        asset_id = self._upload_file(audio_filepath)
-        self.logger.info(f"File uploaded successfully. Asset ID: {asset_id}")
+    api_token: Optional[str] = None
+    base_url: str = "https://groovy.audioshake.ai"
+    output_prefix: Optional[str] = None
+    timeout_minutes: int = 10  # Added timeout configuration
 
-        # Step 2: Create a job for transcription and alignment
-        job_id = self._create_job(asset_id)
-        self.logger.info(f"Job created successfully. Job ID: {job_id}")
 
-        return job_id
+class AudioShakeAPI:
+    """Handles direct API interactions with AudioShake."""
 
-    def get_transcription_result(self, job_id: str) -> dict:
-        """Gets the results for a previously started job."""
-        self.logger.info(f"Getting results for job ID: {job_id}")
+    def __init__(self, config: AudioShakeConfig, logger):
+        self.config = config
+        self.logger = logger
 
-        # Wait for job completion and get results
-        result = self._get_job_result(job_id)
-        self.logger.info(f"Job completed. Processing results...")
+    def _validate_config(self) -> None:
+        """Validate API configuration."""
+        if not self.config.api_token:
+            raise ValueError("AudioShake API token must be provided")
 
-        # Process and return in standard format
-        return self._process_result(result)
+    def _get_headers(self) -> Dict[str, str]:
+        """Get headers for API requests."""
+        self._validate_config()  # Validate before making any API calls
+        return {"Authorization": f"Bearer {self.config.api_token}", "Content-Type": "application/json"}
 
-    def _upload_file(self, filepath):
+    def upload_file(self, filepath: str) -> str:
+        """Upload audio file and return asset ID."""
         self.logger.info(f"Uploading {filepath} to AudioShake")
-        url = f"{self.base_url}/upload"
-        headers = {"Authorization": f"Bearer {self.api_token}"}
+        self._validate_config()  # Validate before making API call
+
+        url = f"{self.config.base_url}/upload"
         with open(filepath, "rb") as file:
             files = {"file": (os.path.basename(filepath), file)}
-            response = requests.post(url, headers=headers, files=files)
-
-        self.logger.info(f"Upload response status code: {response.status_code}")
-        self.logger.info(f"Upload response content: {response.text}")
+            response = requests.post(url, headers={"Authorization": self._get_headers()["Authorization"]}, files=files)
 
+        self.logger.debug(f"Upload response: {response.status_code} - {response.text}")
         response.raise_for_status()
         return response.json()["id"]
 
-    def _create_job(self, asset_id):
+    def create_job(self, asset_id: str) -> str:
+        """Create transcription job and return job ID."""
         self.logger.info(f"Creating job for asset {asset_id}")
-        url = f"{self.base_url}/job/"
-        headers = {"Authorization": f"Bearer {self.api_token}", "Content-Type": "application/json"}
+
+        url = f"{self.config.base_url}/job/"
         data = {
             "metadata": {"format": "json", "name": "alignment", "language": "en"},
             "callbackUrl": "https://example.com/webhook/alignment",
             "assetId": asset_id,
         }
-        response = requests.post(url, headers=headers, json=data)
+        response = requests.post(url, headers=self._get_headers(), json=data)
         response.raise_for_status()
         return response.json()["job"]["id"]
 
-    def _get_job_result(self, job_id):
+    def wait_for_job_result(self, job_id: str) -> Dict[str, Any]:
+        """Poll for job completion and return results."""
         self.logger.info(f"Getting job result for job {job_id}")
-        url = f"{self.base_url}/job/{job_id}"
-        headers = {"Authorization": f"Bearer {self.api_token}", "Content-Type": "application/json"}
+
+        url = f"{self.config.base_url}/job/{job_id}"
+        start_time = time.time()
+        last_status_log = start_time
+        timeout_seconds = self.config.timeout_minutes * 60
+
         while True:
-            response = requests.get(url, headers=headers)
+            current_time = time.time()
+            elapsed_time = current_time - start_time
+
+            # Check for timeout
+            if elapsed_time > timeout_seconds:
+                raise TranscriptionError(f"Transcription timed out after {self.config.timeout_minutes} minutes")
+
+            # Log status every minute
+            if current_time - last_status_log >= 60:
+                self.logger.info(f"Still waiting for transcription... " f"Elapsed time: {int(elapsed_time/60)} minutes")
+                last_status_log = current_time
+
+            response = requests.get(url, headers=self._get_headers())
             response.raise_for_status()
             job_data = response.json()["job"]
+
             if job_data["status"] == "completed":
                 return job_data
             elif job_data["status"] == "failed":
-                raise Exception("Job failed")
-            time.sleep(5)  # Wait 5 seconds before checking again
+                raise TranscriptionError(f"Job failed: {job_data.get('error', 'Unknown error')}")
 
-    def _process_result(self, job_data):
-        self.logger.debug(f"Processing result for job {job_data['id']}")
-        self.logger.debug(f"Job data: {json.dumps(job_data, indent=2)}")
+            time.sleep(5)  # Wait before next poll
 
-        output_assets = job_data.get("outputAssets", [])
-        self.logger.debug(f"Output assets: {output_assets}")
 
-        output_asset = next((asset for asset in output_assets if asset["name"] == "alignment.json"), None)
+class AudioShakeTranscriber(BaseTranscriber):
+    """Transcription service using AudioShake's API."""
 
-        if not output_asset:
-            self.logger.error("'alignment.json' found in job results")
-            self.logger.error(f"Available output assets: {[asset['name'] for asset in output_assets]}")
-            raise Exception("Required output not found in job results")
+    def __init__(
+        self,
+        cache_dir: Union[str, Path],
+        config: Optional[AudioShakeConfig] = None,
+        logger: Optional[Any] = None,
+        api_client: Optional[AudioShakeAPI] = None,
+    ):
+        """Initialize AudioShake transcriber."""
+        super().__init__(cache_dir=cache_dir, logger=logger)
+        self.config = config or AudioShakeConfig(api_token=os.getenv("AUDIOSHAKE_API_TOKEN"))
+        self.api = api_client or AudioShakeAPI(self.config, self.logger)
 
-        transcription_url = output_asset["link"]
-        self.logger.debug(f"Output URL: {transcription_url}")
+    def get_name(self) -> str:
+        return "AudioShake"
 
-        response = requests.get(transcription_url)
-        response.raise_for_status()
-        transcription_data = response.json()
-        self.logger.debug(f"Output data: {json.dumps(transcription_data, indent=2)}")
+    def _perform_transcription(self, audio_filepath: str) -> TranscriptionData:
+        """Actually perform the transcription using AudioShake API."""
+        self.logger.debug(f"Entering _perform_transcription() for {audio_filepath}")
+        self.logger.info(f"Starting transcription for {audio_filepath}")
+
+        try:
+            # Start job and get results
+            self.logger.debug("Calling start_transcription()")
+            job_id = self.start_transcription(audio_filepath)
+            self.logger.debug(f"Got job_id: {job_id}")
+
+            self.logger.debug("Calling get_transcription_result()")
+            result = self.get_transcription_result(job_id)
+            self.logger.debug("Got transcription result")
 
-        transcription_data = {"segments": transcription_data.get("lines", []), "text": transcription_data.get("text", "")}
+            return result
+        except Exception as e:
+            self.logger.error(f"Error in _perform_transcription: {str(e)}")
+            raise
 
-        # Ensure each segment has the required fields
-        for segment in transcription_data["segments"]:
-            if "words" not in segment:
-                segment["words"] = []
-            if "text" not in segment:
-                segment["text"] = " ".join(word["text"] for word in segment["words"])
+    def start_transcription(self, audio_filepath: str) -> str:
+        """Starts the transcription job and returns the job ID."""
+        self.logger.debug(f"Entering start_transcription() for {audio_filepath}")
+
+        # Upload file and create job
+        asset_id = self.api.upload_file(audio_filepath)
+        self.logger.debug(f"File uploaded successfully. Asset ID: {asset_id}")
+
+        job_id = self.api.create_job(asset_id)
+        self.logger.debug(f"Job created successfully. Job ID: {job_id}")
+
+        return job_id
+
+    def get_transcription_result(self, job_id: str) -> Dict[str, Any]:
+        """Gets the raw results for a previously started job."""
+        self.logger.debug(f"Entering get_transcription_result() for job ID: {job_id}")
 
-        transcription_data["output_filename"] = self.get_output_filename(" (AudioShake)")
+        # Wait for job completion
+        job_data = self.api.wait_for_job_result(job_id)
+        self.logger.debug("Job completed. Getting results...")
 
-        return transcription_data
+        output_asset = next((asset for asset in job_data.get("outputAssets", []) if asset["name"] == "alignment.json"), None)
+        if not output_asset:
+            raise TranscriptionError("Required output not found in job results")
+
+        # Fetch transcription data
+        response = requests.get(output_asset["link"])
+        response.raise_for_status()
 
-    def get_output_filename(self, suffix):
-        """Generate consistent filename with (Purpose) suffix pattern"""
-        return f"{self.output_prefix}{suffix}"
+        # Return combined raw data
+        raw_data = {"job_data": job_data, "transcription": response.json()}
+
+        self.logger.debug("Raw results retrieved successfully")
+        return raw_data
+
+    def _convert_result_format(self, raw_data: Dict[str, Any]) -> TranscriptionData:
+        """Process raw Audioshake API response into standard format."""
+        self.logger.debug(f"Processing result for job {raw_data['job_data']['id']}")
+
+        transcription_data = raw_data["transcription"]
+        job_data = raw_data["job_data"]
+
+        segments = []
+        all_words = []  # Collect all words across segments
+
+        for line in transcription_data.get("lines", []):
+            words = [
+                Word(
+                    text=word["text"].strip(" "),
+                    start_time=word.get("start", 0.0),
+                    end_time=word.get("end", 0.0),
+                )
+                for word in line.get("words", [])
+            ]
+            all_words.extend(words)  # Add words to flat list
+
+            segments.append(
+                LyricsSegment(
+                    text=line.get("text", " ".join(w.text for w in words)),
+                    words=words,
+                    start_time=min((w.start_time for w in words), default=0.0),
+                    end_time=max((w.end_time for w in words), default=0.0),
+                )
+            )
+
+        return TranscriptionData(
+            text=transcription_data.get("text", ""),
+            words=all_words,
+            segments=segments,
+            source=self.get_name(),
+            metadata={
+                "language": transcription_data.get("metadata", {}).get("language"),
+                "job_id": job_data["id"],
+                "duration": job_data.get("statusInfo", {}).get("duration"),
+            },
+        )
+
+    def get_output_filename(self, suffix: str) -> str:
+        """Generate consistent filename with (Purpose) suffix pattern."""
+        return f"{self.config.output_prefix}{suffix}"

lyrics_transcriber/transcribers/base_transcriber.py

@@ -0,0 +1,149 @@
+from abc import ABC, abstractmethod
+from typing import Dict, Any, Optional, Union
+from pathlib import Path
+import logging
+import os
+import json
+import hashlib
+from lyrics_transcriber.types import TranscriptionData
+
+
+class TranscriptionError(Exception):
+    """Base exception for transcription errors."""
+
+    def __init__(self, message: str):
+        super().__init__(message)
+
+
+class BaseTranscriber(ABC):
+    """Base class for all transcription services."""
+
+    def __init__(self, cache_dir: Union[str, Path], logger: Optional[logging.Logger] = None):
+        """
+        Initialize transcriber with cache directory and logger.
+
+        Args:
+            cache_dir: Directory to store cache files. Must be provided.
+            logger: Logger instance to use. If None, creates a new logger.
+        """
+        self.cache_dir = Path(cache_dir)
+        self.logger = logger or logging.getLogger(__name__)
+
+        self.cache_dir.mkdir(parents=True, exist_ok=True)
+        self.logger.debug(f"Initialized {self.__class__.__name__} with cache dir: {self.cache_dir}")
+
+    def _get_file_hash(self, filepath: str) -> str:
+        """Calculate MD5 hash of a file."""
+        self.logger.debug(f"Calculating hash for file: {filepath}")
+        md5_hash = hashlib.md5()
+        with open(filepath, "rb") as f:
+            for chunk in iter(lambda: f.read(4096), b""):
+                md5_hash.update(chunk)
+        hash_result = md5_hash.hexdigest()
+        self.logger.debug(f"File hash: {hash_result}")
+        return hash_result
+
+    def _get_cache_path(self, file_hash: str, suffix: str) -> str:
+        """Get the cache file path for a given file hash."""
+        cache_path = os.path.join(self.cache_dir, f"{self.get_name().lower()}_{file_hash}_{suffix}.json")
+        self.logger.debug(f"Cache path: {cache_path}")
+        return cache_path
+
+    def _save_to_cache(self, cache_path: str, raw_data: Dict[str, Any]) -> None:
+        """Save raw API response data to cache."""
+        self.logger.debug(f"Saving JSON to cache: {cache_path}")
+        with open(cache_path, "w") as f:
+            json.dump(raw_data, f, indent=2)
+        self.logger.debug("Cache save completed")
+
+    def _load_from_cache(self, cache_path: str) -> Optional[Dict[str, Any]]:
+        """Load raw API response data from cache if it exists."""
+        self.logger.debug(f"Attempting to load from cache: {cache_path}")
+        try:
+            with open(cache_path, "r") as f:
+                data = json.load(f)
+                self.logger.debug("Raw API response loaded from cache")
+                return data
+        except FileNotFoundError:
+            self.logger.debug("Cache file not found")
+            return None
+        except json.JSONDecodeError:
+            self.logger.warning(f"Cache file {cache_path} is corrupted")
+            return None
+
+    def _save_and_convert_result(self, file_hash: str, raw_result: Dict[str, Any]) -> TranscriptionData:
+        """Convert raw result to TranscriptionData, save to cache, and return."""
+        converted_cache_path = self._get_cache_path(file_hash, "converted")
+        converted_result = self._convert_result_format(raw_result)
+        self._save_to_cache(converted_cache_path, converted_result.to_dict())
+        return converted_result
+
+    def transcribe(self, audio_filepath: str) -> TranscriptionData:
+        """
+        Transcribe an audio file, using cache if available.
+
+        Args:
+            audio_filepath: Path to the audio file to transcribe
+
+        Returns:
+            TranscriptionData containing segments, text, and metadata
+        """
+        self.logger.debug(f"Starting transcription for {audio_filepath}")
+
+        try:
+            self._validate_audio_file(audio_filepath)
+            self.logger.debug("Audio file validation passed")
+
+            # Check cache first
+            file_hash = self._get_file_hash(audio_filepath)
+            raw_cache_path = self._get_cache_path(file_hash, "raw")
+
+            raw_data = self._load_from_cache(raw_cache_path)
+            if raw_data:
+                self.logger.info(f"Using cached raw data for {audio_filepath}")
+                return self._save_and_convert_result(file_hash, raw_data)
+
+            # If not in cache, perform transcription
+            self.logger.info(f"No cache found, transcribing {audio_filepath}")
+            raw_result = self._perform_transcription(audio_filepath)
+            self.logger.debug("Transcription completed")
+
+            # Save raw result to cache
+            self._save_to_cache(raw_cache_path, raw_result)
+
+            return self._save_and_convert_result(file_hash, raw_result)
+
+        except Exception as e:
+            self.logger.error(f"Error during transcription: {str(e)}")
+            raise
+
+    @abstractmethod
+    def _perform_transcription(self, audio_filepath: str) -> TranscriptionData:
+        """
+        Actually perform the transcription (implemented by subclasses).
+
+        Args:
+            audio_filepath: Path to the audio file to transcribe
+
+        Returns:
+            TranscriptionData containing segments, text, and metadata
+        """
+        pass  # pragma: no cover
+
+    @abstractmethod
+    def get_name(self) -> str:
+        """Return the name of this transcription service."""
+        pass  # pragma: no cover
+
+    def _validate_audio_file(self, audio_filepath: str) -> None:
+        """Validate that the audio file exists and is accessible."""
+        self.logger.debug(f"Validating audio file: {audio_filepath}")
+        if not os.path.exists(audio_filepath):
+            self.logger.error(f"Audio file not found: {audio_filepath}")
+            raise FileNotFoundError(f"Audio file not found: {audio_filepath}")
+        self.logger.debug("Audio file validation successful")
+
+    @abstractmethod
+    def _convert_result_format(self, raw_data: Dict[str, Any]) -> TranscriptionData:
+        """Convert raw API response to TranscriptionData format."""
+        pass  # pragma: no cover