Add AssemblyAI studio flow and align versions

.github/workflows/release.yml

@@ -99,6 +99,7 @@ jobs:
           node-version: '20'
       - name: Build studio bundle
         run: |
+          npm version --no-git-tag-version --allow-same-version "${GITHUB_REF_NAME#v}"
           npm ci
           sh scripts/build-studio.sh dist/studio
           tar -czf studio-bundle.tar.gz -C dist/studio .

backend/cli.py

@@ -18,6 +18,7 @@
 import sys
 import textwrap
 import time
+from version import VERSION
 
 # Windows stdout/stderr default to cp1252, which can't encode chars like '→'; output is UTF-8.
 for _stream in (sys.stdout, sys.stderr):
@@ -360,6 +361,11 @@ def cmd_studio(args):
         cmd += ["--paragraph", args.paragraph]
     if args.language:
         cmd += ["--language", args.language]
+    if getattr(args, "engine", None):
+        cmd += ["--engine", args.engine]
+    env = os.environ.copy()
+    if getattr(args, "assemblyai_api_key", None):
+        env["ASSEMBLYAI_API_KEY"] = args.assemblyai_api_key
     cmd += [
         "--caption-style", args.caption_style,
         "--crop", args.crop,
@@ -386,7 +392,7 @@ def cmd_studio(args):
         cmd += ["--no-outro"]
 
     import subprocess
-    rc = subprocess.run(cmd).returncode
+    rc = subprocess.run(cmd, env=env).returncode
     sys.exit(rc)
 
 
@@ -437,6 +443,8 @@ def cmd_process(args):
     # CLI overrides
     if getattr(args, "engine", None):
         os.environ["PODCLI_ENGINE"] = args.engine
+    if getattr(args, "assemblyai_api_key", None):
+        os.environ["ASSEMBLYAI_API_KEY"] = args.assemblyai_api_key
     if args.caption_style:
         config["caption_style"] = args.caption_style
     if args.crop:
@@ -585,7 +593,9 @@ def cmd_process(args):
             result = cached
             print(f"         {len(segments)} segments, {len(words)} words")
         else:
-            print("  [1/4] Transcribing with Whisper...")
+            from services.engines import is_assemblyai_engine
+            engine_label = "AssemblyAI" if is_assemblyai_engine(os.environ.get("PODCLI_ENGINE", "")) else "Whisper"
+            print(f"  [1/4] Transcribing with {engine_label}...")
             _ensure_ssl_certs()
             import warnings
             warnings.filterwarnings("ignore", message="FP16 is not supported on CPU")
@@ -615,6 +625,7 @@ def _transcribe_progress(pct, msg):
                 result = transcribe_file(
                     file_path=video_path,
                     model_size=config.get("whisper_model", "base"),
+                    engine=os.environ.get("PODCLI_ENGINE") or None,
                     enable_diarization=not config.get("no_speakers", False),
                     progress_callback=_transcribe_progress,
                 )
@@ -3095,8 +3106,6 @@ def cmd_info(args):
     print()
 
 
-VERSION = "1.0.0"
-
 BANNER = """
 \033[38;2;212;135;74m  ┌─────────────────────────────────────┐
   │                                     │
@@ -3309,7 +3318,8 @@ def main():
     proc.add_argument("-n", "--top", type=int, help="Number of top clips to export (default: 5)")
     proc.add_argument("-o", "--output", help="Output directory (default: ./clips)")
     proc.add_argument("-p", "--preset", help="Load a saved preset")
-    proc.add_argument("--engine", choices=["whisper-py", "whispercpp"], help="Transcription engine (default: whisper-py; whispercpp is the native, PyTorch-free path)")
+    proc.add_argument("--engine", choices=["whisper-py", "whispercpp", "assemblyai"], help="Transcription engine (default: whisper-py; whispercpp is local; assemblyai uses ASSEMBLYAI_API_KEY)")
+    proc.add_argument("--assemblyai-api-key", help="AssemblyAI API key for --engine assemblyai. Prefer ASSEMBLYAI_API_KEY; command-line secrets can appear in process listings.")
     proc.add_argument("--fast", action="store_true", help="Draft mode: tiny Whisper, heuristic selection, center crop, low quality")
     proc.add_argument("--thumbnails", dest="thumbnails", action="store_true", default=None, help="Force thumbnail generation on")
     proc.add_argument("--no-thumbnails", dest="thumbnails", action="store_false", help="Skip thumbnail generation")
@@ -3335,6 +3345,8 @@ def main():
     studio.add_argument("--end", type=float, help="Fragment end (seconds)")
     studio.add_argument("--paragraph", help="Find the fragment by matching this text in the transcript")
     studio.add_argument("--language", help="Transcription language (e.g. es). Auto-detect if omitted.")
+    studio.add_argument("--engine", choices=["whisper-py", "whispercpp", "assemblyai"], help="Transcription engine")
+    studio.add_argument("--assemblyai-api-key", help="AssemblyAI API key for --engine assemblyai. Prefer ASSEMBLYAI_API_KEY; command-line secrets can appear in process listings.")
     studio.add_argument("--caption-style", choices=["hormozi", "karaoke", "subtle", "branded"], default="hormozi")
     studio.add_argument("--crop", choices=["center", "face", "speaker", "speaker-hardcut"], default="face")
     studio.add_argument("-o", "--output", help="Final output path")

backend/clip_studio.py

@@ -82,12 +82,14 @@ def _probe_duration(path: str) -> float:
         return 0.0
 
 
-def _transcribe(video: str, language: str | None):
+def _transcribe(video: str, language: str | None, engine: str | None):
     """Transcribe (cached) and return the word list."""
+    from services.engines import is_assemblyai_engine
     from services.transcription import transcribe_file
-    print("  [transcribe] running Whisper...", flush=True)
+    label = "AssemblyAI" if is_assemblyai_engine(engine or os.environ.get("PODCLI_ENGINE", "")) else "Whisper"
+    print(f"  [transcribe] running {label}...", flush=True)
     res = transcribe_file(
-        file_path=video, model_size="base", language=language,
+        file_path=video, model_size="base", engine=engine, language=language,
         enable_diarization=False,
         progress_callback=lambda p, m: print(f"    {p}% {m}", flush=True),
     )
@@ -199,6 +201,7 @@ def main():
     ap.add_argument("--end", type=float, help="Fragment end (seconds)")
     ap.add_argument("--paragraph", help="Find fragment by matching this text in the transcript")
     ap.add_argument("--language", default=None, help="Transcription language (e.g. es). Auto-detect if omitted.")
+    ap.add_argument("--engine", choices=["whisper-py", "whispercpp", "assemblyai"], default=None, help="Transcription engine")
     ap.add_argument("--caption-style", default="hormozi", choices=["hormozi", "karaoke", "subtle", "branded"])
     ap.add_argument("--crop", default="face", choices=["center", "face", "speaker", "speaker-hardcut"])
     ap.add_argument("--output", default=None, help="Final output path")
@@ -244,7 +247,7 @@ def main():
     os.makedirs(out_dir, exist_ok=True)
 
     # Need a transcript if cutting by paragraph or if rendering captions.
-    words = _transcribe(video, args.language)
+    words = _transcribe(video, args.language, args.engine)
 
     if args.paragraph:
         start, end = _find_paragraph(words, args.paragraph)

backend/main.py

@@ -25,6 +25,8 @@
 except ImportError:
     pass
 import traceback
+from version import VERSION
+
 
 def emit_progress(task_id: str, stage: str, percent: int, message: str, **extra):
     """Write a progress event to stderr (picked up by TypeScript executor)."""
@@ -53,25 +55,43 @@ def emit_result(task_id: str, status: str, data=None, error=None):
 
 def handle_ping(task_id: str, params: dict):
     """Simple health check."""
-    emit_result(task_id, "success", data={"message": "pong", "version": "1.0.0"})
+    emit_result(task_id, "success", data={"message": "pong", "version": VERSION})
 
 
 def handle_transcribe(task_id: str, params: dict):
     """Transcribe a podcast video/audio file with speaker detection."""
     from services.transcription import transcribe_file
     from services.corrections import apply_corrections
-    from services.transcript_packer import compute_cache_hash, write_packed
+    from services.transcript_packer import compute_cache_hash, engine_cache_suffix, write_packed
 
     emit_progress(task_id, "transcribing", 0, "Starting transcription...")
     file_path = params["file_path"]
-    result = transcribe_file(
-        file_path=file_path,
-        model_size=params.get("model_size", "base"),
-        language=params.get("language"),
-        enable_diarization=params.get("enable_diarization", True),
-        num_speakers=params.get("num_speakers"),
-        progress_callback=lambda pct, msg: emit_progress(task_id, "transcribing", pct, msg),
-    )
+    engine = params.get("engine")
+    previous_engine = os.environ.get("PODCLI_ENGINE")
+    previous_assemblyai_key = os.environ.get("ASSEMBLYAI_API_KEY")
+    if engine:
+        os.environ["PODCLI_ENGINE"] = engine
+    if params.get("assemblyai_api_key"):
+        os.environ["ASSEMBLYAI_API_KEY"] = params["assemblyai_api_key"]
+    try:
+        result = transcribe_file(
+            file_path=file_path,
+            model_size=params.get("model_size", "base"),
+            engine=engine,
+            language=params.get("language"),
+            enable_diarization=params.get("enable_diarization", True),
+            num_speakers=params.get("num_speakers"),
+            progress_callback=lambda pct, msg: emit_progress(task_id, "transcribing", pct, msg),
+        )
+    finally:
+        if previous_engine is None:
+            os.environ.pop("PODCLI_ENGINE", None)
+        else:
+            os.environ["PODCLI_ENGINE"] = previous_engine
+        if previous_assemblyai_key is None:
+            os.environ.pop("ASSEMBLYAI_API_KEY", None)
+        else:
+            os.environ["ASSEMBLYAI_API_KEY"] = previous_assemblyai_key
     # Apply word corrections (Whisper misheard proper nouns)
     apply_corrections(result.get("words", []), result.get("segments", []))
 
@@ -80,7 +100,7 @@ def handle_transcribe(task_id: str, params: dict):
     try:
         from services.audio_analyzer import extract_audio_energy
 
-        cache_hash = compute_cache_hash(file_path)
+        cache_hash = compute_cache_hash(file_path) + engine_cache_suffix(result.get("engine") or engine)
         energy_data = None
         try:
             energy_data = extract_audio_energy(file_path)

backend/services/content_generator.py

@@ -7,6 +7,7 @@
 import json
 import os
 import subprocess
+import sys
 import tempfile
 import threading
 from typing import Optional, Callable
@@ -234,7 +235,8 @@ def generate_custom_content(
                     project_dir=project_dir,
                     timeout=120,
                 )
-            except Exception:
+            except Exception as exc:
+                print(f"Warning: {label} content generation failed: {exc}", file=sys.stderr)
                 continue
             if cr.returncode != 0 or not cr.stdout.strip():
                 continue
@@ -245,8 +247,8 @@ def generate_custom_content(
     finally:
         try:
             os.unlink(prompt_file)
-        except Exception:
-            pass
+        except Exception as exc:
+            print(f"Warning: could not remove prompt file {prompt_file}: {exc}", file=sys.stderr)
 
 
 def generate_clip_content(

backend/services/engines.py

@@ -0,0 +1,11 @@
+def normalize_engine(name: str | None) -> str:
+    value = (name or "whisper-py").strip().lower()
+    if value in ("whispercpp", "whisper-cpp", "whisper.cpp", "cpp"):
+        return "whispercpp"
+    if value in ("assemblyai", "assembly-ai", "aai"):
+        return "assemblyai"
+    return "whisper-py"
+
+
+def is_assemblyai_engine(name: str | None) -> bool:
+    return normalize_engine(name) == "assemblyai"

backend/services/env_settings.py

@@ -18,6 +18,14 @@
         "url": "https://huggingface.co/settings/tokens",
         "secret": True,
     },
+    {
+        "key": "ASSEMBLYAI_API_KEY",
+        "label": "AssemblyAI API key",
+        "help": "Enables AssemblyAI transcription in the Studio and CLI.",
+        "url": "https://www.assemblyai.com/dashboard/activation",
+        "secret": True,
+        "placeholder": "aai_...",
+    },
     {
         "key": "PODCLI_CLAUDE_PATH",
         "label": "Claude Code CLI path",

backend/services/transcript_packer.py

@@ -19,6 +19,7 @@
 from typing import Any, Optional
 
 from config.paths import paths
+from services.engines import normalize_engine
 
 def _transcripts_cache_dir() -> str:
     return paths["transcripts"]
@@ -66,9 +67,15 @@ def _engine_cache_suffix() -> str:
     """Namespace the cache by engine so a whisper.cpp run doesn't reuse a
     whisper-py transcript (their timings/word splits differ). whisper-py keeps
     the bare filename, which the TS transcript cache also writes."""
-    engine = os.environ.get("PODCLI_ENGINE", "whisper-py").strip().lower()
-    if engine in ("whispercpp", "whisper-cpp", "whisper.cpp", "cpp"):
+    return engine_cache_suffix(os.environ.get("PODCLI_ENGINE", "whisper-py"))
+
+
+def engine_cache_suffix(engine: str | None) -> str:
+    engine = normalize_engine(engine)
+    if engine == "whispercpp":
         return "-whispercpp"
+    if engine == "assemblyai":
+        return "-assemblyai"
     return ""