seadanceai-transcript-sdk 0.1.0__tar.gz

seadanceai_transcript_sdk-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Seadance AI
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

seadanceai_transcript_sdk-0.1.0/PKG-INFO

@@ -0,0 +1,93 @@
+Metadata-Version: 2.4
+Name: seadanceai-transcript-sdk
+Version: 0.1.0
+Summary: Python toolkit sponsored by Seadance AI for fetching YouTube transcripts, languages, and lightweight summaries.
+Author: Seadance AI
+License-Expression: MIT
+Project-URL: Homepage, https://seadanceai.com/?ref=seadanceai-transcript-sdk-py
+Project-URL: Repository, https://github.com/zeikia/seadanceai-transcript-sdk
+Project-URL: Sponsor, https://seadanceai.com/?ref=seadanceai-transcript-sdk-py
+Keywords: youtube,transcript,seadance,ai
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: requests>=2.31
+Requires-Dist: youtube-transcript-api>=0.6.2
+Dynamic: license-file
+
+# seadanceai-transcript-sdk (Python)
+
+[![PyPI](https://img.shields.io/pypi/v/seadanceai-transcript-sdk.svg?color=0aa&style=flat-square)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![Python](https://img.shields.io/pypi/pyversions/seadanceai-transcript-sdk.svg?style=flat-square&color=3776ab)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![License](https://img.shields.io/pypi/l/seadanceai-transcript-sdk.svg?style=flat-square&color=facc15)](LICENSE)
+[![Sponsored by Seadance AI](https://img.shields.io/badge/Sponsored%20by-Seadance%20AI-f97316?style=flat-square)](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py)
+
+Community-maintained Python toolkit for retrieving YouTube transcripts, checking available languages, exporting TXT/SRT/VTT files, and generating heuristic summaries. Proudly sponsored by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py) — mention them or link back if this library helps your product!
+
+## ✨ Features
+
+- **Dual-source transcripts**: tactiq endpoint first, fallback to `youtube-transcript-api` for reliability.
+- **Language probing**: batch-check common caption languages.
+- **Formatters**: output transcripts as TXT/SRT/VTT with helper functions.
+- **Summary helpers**: generate concise recaps + key points without external APIs.
+- **Zero external API keys** required; just Python + HTTP.
+
+## 📦 Installation
+
+```bash
+pip install seadanceai-transcript-sdk
+```
+
+## ⚡ Quick Start
+
+```python
+from seadance_transcript_sdk import TranscriptClient, format_transcript_file
+
+client = TranscriptClient(enable_fallback=True)
+transcript = client.get_transcript("dQw4w9WgXcQ", language="en")
+file_payload = format_transcript_file(transcript, fmt="srt")
+with open(file_payload.filename, "w", encoding="utf-8") as fh:
+    fh.write(file_payload.content)
+```
+
+More helpers:
+
+```python
+from seadance_transcript_sdk import detect_available_languages
+
+langs = detect_available_languages("dQw4w9WgXcQ")
+print([lang.name for lang in langs.languages])
+
+summary = client.summarize(transcript)
+print(summary.summary)
+```
+
+## 🧩 Modules
+
+- `seadance_transcript_sdk.transcript` – fetch logic + language probing.
+- `seadance_transcript_sdk.formatting` – TXT/SRT/VTT helpers.
+- `seadance_transcript_sdk.summary` – heuristic summary utilities.
+- `seadance_transcript_sdk.client` – ergonomic wrapper with sensible defaults.
+
+## 🤝 Contributing
+
+1. `python -m venv .venv && source .venv/bin/activate`
+2. `pip install -e .[dev]` (dev extras coming soon) or simply `pip install -r requirements-dev.txt`
+3. `pytest` (todo) then `python -m build`
+4. `twine upload dist/*`
+
+## 🙌 Sponsorship
+
+Development is powered by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py). Link back to them or reach out if you want bespoke media-intelligence tooling.
+
+## 📄 License
+
+MIT © Seadance AI & contributors.

seadanceai_transcript_sdk-0.1.0/README.md

@@ -0,0 +1,68 @@
+# seadanceai-transcript-sdk (Python)
+
+[![PyPI](https://img.shields.io/pypi/v/seadanceai-transcript-sdk.svg?color=0aa&style=flat-square)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![Python](https://img.shields.io/pypi/pyversions/seadanceai-transcript-sdk.svg?style=flat-square&color=3776ab)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![License](https://img.shields.io/pypi/l/seadanceai-transcript-sdk.svg?style=flat-square&color=facc15)](LICENSE)
+[![Sponsored by Seadance AI](https://img.shields.io/badge/Sponsored%20by-Seadance%20AI-f97316?style=flat-square)](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py)
+
+Community-maintained Python toolkit for retrieving YouTube transcripts, checking available languages, exporting TXT/SRT/VTT files, and generating heuristic summaries. Proudly sponsored by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py) — mention them or link back if this library helps your product!
+
+## ✨ Features
+
+- **Dual-source transcripts**: tactiq endpoint first, fallback to `youtube-transcript-api` for reliability.
+- **Language probing**: batch-check common caption languages.
+- **Formatters**: output transcripts as TXT/SRT/VTT with helper functions.
+- **Summary helpers**: generate concise recaps + key points without external APIs.
+- **Zero external API keys** required; just Python + HTTP.
+
+## 📦 Installation
+
+```bash
+pip install seadanceai-transcript-sdk
+```
+
+## ⚡ Quick Start
+
+```python
+from seadance_transcript_sdk import TranscriptClient, format_transcript_file
+
+client = TranscriptClient(enable_fallback=True)
+transcript = client.get_transcript("dQw4w9WgXcQ", language="en")
+file_payload = format_transcript_file(transcript, fmt="srt")
+with open(file_payload.filename, "w", encoding="utf-8") as fh:
+    fh.write(file_payload.content)
+```
+
+More helpers:
+
+```python
+from seadance_transcript_sdk import detect_available_languages
+
+langs = detect_available_languages("dQw4w9WgXcQ")
+print([lang.name for lang in langs.languages])
+
+summary = client.summarize(transcript)
+print(summary.summary)
+```
+
+## 🧩 Modules
+
+- `seadance_transcript_sdk.transcript` – fetch logic + language probing.
+- `seadance_transcript_sdk.formatting` – TXT/SRT/VTT helpers.
+- `seadance_transcript_sdk.summary` – heuristic summary utilities.
+- `seadance_transcript_sdk.client` – ergonomic wrapper with sensible defaults.
+
+## 🤝 Contributing
+
+1. `python -m venv .venv && source .venv/bin/activate`
+2. `pip install -e .[dev]` (dev extras coming soon) or simply `pip install -r requirements-dev.txt`
+3. `pytest` (todo) then `python -m build`
+4. `twine upload dist/*`
+
+## 🙌 Sponsorship
+
+Development is powered by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py). Link back to them or reach out if you want bespoke media-intelligence tooling.
+
+## 📄 License
+
+MIT © Seadance AI & contributors.

seadanceai_transcript_sdk-0.1.0/pyproject.toml

@@ -0,0 +1,43 @@
+[build-system]
+requires = ["setuptools>=69", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "seadanceai-transcript-sdk"
+version = "0.1.0"
+description = "Python toolkit sponsored by Seadance AI for fetching YouTube transcripts, languages, and lightweight summaries."
+readme = "README.md"
+authors = [
+  { name = "Seadance AI" }
+]
+license = "MIT"
+requires-python = ">=3.9"
+dependencies = [
+  "requests>=2.31",
+  "youtube-transcript-api>=0.6.2"
+]
+keywords = ["youtube", "transcript", "seadance", "ai"]
+classifiers = [
+  "Development Status :: 3 - Alpha",
+  "Intended Audience :: Developers",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3.9",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
+  "Operating System :: OS Independent"
+]
+
+[project.urls]
+Homepage = "https://seadanceai.com/?ref=seadanceai-transcript-sdk-py"
+Repository = "https://github.com/zeikia/seadanceai-transcript-sdk"
+Sponsor = "https://seadanceai.com/?ref=seadanceai-transcript-sdk-py"
+
+[tool.setuptools]
+package-dir = { "" = "src" }
+include-package-data = true
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+

seadanceai_transcript_sdk-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/__init__.py

@@ -0,0 +1,29 @@
+from .client import TranscriptClient
+from .formatting import format_transcript, format_transcript_file
+from .summary import summarize_transcript
+from .transcript import detect_available_languages, get_transcript
+from .types import (
+  LanguageOption,
+  LanguagesResponse,
+  SummaryQuestion,
+  SummaryResult,
+  TranscriptFormatResult,
+  TranscriptResponse,
+  TranscriptSnippet,
+)
+
+__all__ = [
+  "TranscriptClient",
+  "get_transcript",
+  "detect_available_languages",
+  "format_transcript",
+  "format_transcript_file",
+  "summarize_transcript",
+  "TranscriptResponse",
+  "TranscriptSnippet",
+  "LanguagesResponse",
+  "LanguageOption",
+  "SummaryResult",
+  "SummaryQuestion",
+  "TranscriptFormatResult",
+]

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/client.py

@@ -0,0 +1,33 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import List, Optional
+
+from .formatting import format_transcript, format_transcript_file
+from .summary import summarize_transcript
+from .transcript import detect_available_languages, get_transcript
+from .types import LanguagesResponse, SummaryResult, TranscriptFormatResult, TranscriptResponse
+
+
+@dataclass
+class TranscriptClient:
+  enable_fallback: bool = True
+  tactiq_endpoint: str | None = None
+
+  def get_transcript(self, video_id: str, language: str = "en", *, enable_fallback: Optional[bool] = None) -> TranscriptResponse:
+    return get_transcript(video_id, language=language, enable_fallback=self._resolve_fallback(enable_fallback), tactiq_endpoint=self.tactiq_endpoint or None)
+
+  def detect_languages(self, video_id: str) -> LanguagesResponse:
+    return detect_available_languages(video_id)
+
+  def summarize(self, transcript: TranscriptResponse, title: str | None = None) -> SummaryResult:
+    return summarize_transcript(transcript, title=title)
+
+  def format(self, transcript: TranscriptResponse, fmt: str = "txt") -> str:
+    return format_transcript(transcript, fmt)
+
+  def format_to_file(self, transcript: TranscriptResponse, fmt: str = "txt") -> TranscriptFormatResult:
+    return format_transcript_file(transcript, fmt)
+
+  def _resolve_fallback(self, override: Optional[bool]) -> bool:
+    return self.enable_fallback if override is None else override

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/constants.py

@@ -0,0 +1,25 @@
+DEFAULT_TACTIQ_ENDPOINT = "https://tactiq-apps-prod.tactiq.io/transcript"
+DEFAULT_LANGUAGE = "en"
+YOUTUBE_URL_PREFIX = "https://www.youtube.com/watch?v="
+
+LANGUAGE_DISPLAY_NAMES = {
+  "en": "English",
+  "es": "Spanish",
+  "fr": "French",
+  "de": "German",
+  "ja": "Japanese",
+  "ko": "Korean",
+  "zh": "Chinese",
+  "pt": "Portuguese",
+  "ru": "Russian",
+  "ar": "Arabic",
+  "hi": "Hindi",
+  "it": "Italian",
+  "nl": "Dutch",
+  "pl": "Polish",
+  "tr": "Turkish",
+  "th": "Thai",
+  "vi": "Vietnamese"
+}
+
+DEFAULT_LANGUAGE_PROBE_ORDER = list(LANGUAGE_DISPLAY_NAMES.keys())

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/formatting.py

@@ -0,0 +1,60 @@
+from __future__ import annotations
+
+from .types import TranscriptFormatResult, TranscriptResponse
+
+MIME_TYPES = {
+  "txt": "text/plain; charset=utf-8",
+  "srt": "application/x-subrip",
+  "vtt": "text/vtt"
+}
+
+
+def format_transcript(transcript: TranscriptResponse, fmt: str = "txt") -> str:
+  fmt = fmt.lower()
+  if fmt == "srt":
+    return _format_srt(transcript)
+  if fmt == "vtt":
+    return _format_vtt(transcript)
+  return _format_txt(transcript)
+
+
+def format_transcript_file(transcript: TranscriptResponse, fmt: str = "txt") -> TranscriptFormatResult:
+  fmt = fmt.lower()
+  content = format_transcript(transcript, fmt)
+  filename = f"transcript-{transcript.video_id}.{fmt}"
+  mime_type = MIME_TYPES.get(fmt, MIME_TYPES["txt"])
+  return TranscriptFormatResult(content=content, filename=filename, mime_type=mime_type)
+
+
+def _pad(value: int, size: int = 2) -> str:
+  return str(value).zfill(size)
+
+
+def _format_timestamp(seconds: float) -> str:
+  hours = int(seconds // 3600)
+  minutes = int((seconds % 3600) // 60)
+  secs = int(seconds % 60)
+  millis = int((seconds % 1) * 1000)
+  return f"{_pad(hours)}:{_pad(minutes)}:{_pad(secs)},{_pad(millis, 3)}"
+
+
+def _format_srt(transcript: TranscriptResponse) -> str:
+  lines = []
+  for index, snippet in enumerate(transcript.snippets, start=1):
+    start = _format_timestamp(snippet.start)
+    end = _format_timestamp(snippet.start + snippet.duration)
+    lines.append(f"{index}\n{start} --> {end}\n{snippet.text}\n")
+  return "\n".join(lines).strip()
+
+
+def _format_vtt(transcript: TranscriptResponse) -> str:
+  entries = []
+  for snippet in transcript.snippets:
+    start = _format_timestamp(snippet.start)
+    end = _format_timestamp(snippet.start + snippet.duration)
+    entries.append(f"{start} --> {end}\n{snippet.text}")
+  return "WEBVTT\n\n" + "\n\n".join(entries)
+
+
+def _format_txt(transcript: TranscriptResponse) -> str:
+  return "\n".join(snippet.text for snippet in transcript.snippets)

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/summary.py

@@ -0,0 +1,62 @@
+from __future__ import annotations
+
+from typing import Iterable, List
+
+from .types import SummaryQuestion, SummaryResult, TranscriptResponse
+
+
+def summarize_transcript(transcript: TranscriptResponse, title: str | None = None) -> SummaryResult:
+  snippets = transcript.snippets
+  if not snippets:
+    return SummaryResult(summary="No transcript available.")
+
+  full_text = " ".join(snippet.text for snippet in snippets)
+  sentences = [sentence.strip() for sentence in full_text.split('.') if len(sentence.strip()) > 10]
+
+  interval = max(1, len(sentences) // 5) if sentences else 1
+  key_sentences: List[str] = []
+  keywords = ["important", "key", "main", "first", "second", "finally", "conclusion", "summary"]
+  for idx, sentence in enumerate(sentences):
+    lower = sentence.lower()
+    if any(keyword in lower for keyword in keywords) or idx % interval == 0:
+      key_sentences.append(sentence)
+    if len(key_sentences) >= 8:
+      break
+
+  summary_text = ". ".join(key_sentences) + "." if key_sentences else full_text[:200] + "..."
+  key_points = _extract_key_points(snippets)
+  questions = _build_top_questions(title or f"YouTube Video {transcript.video_id}")
+  duration = _estimate_duration(snippets)
+
+  return SummaryResult(summary=summary_text.strip(), key_points=key_points, top_questions=questions, duration_seconds=duration)
+
+
+def _extract_key_points(snippets: Iterable) -> List[str]:
+  snippets = list(snippets)
+  if not snippets:
+    return ["No key points available"]
+
+  interval = max(1, len(snippets) // 6)
+  points: List[str] = []
+  for index in range(0, len(snippets), interval):
+    text = snippets[index].text.strip()
+    if len(text) > 15:
+      snippet = text[:120] + ("..." if len(text) > 120 else "")
+      points.append(snippet)
+  return points or ["Key points derived from transcript"]
+
+
+def _build_top_questions(title: str) -> List[SummaryQuestion]:
+  lower_title = title.lower()
+  return [
+    SummaryQuestion(question=f"What is the main topic of '{title}'?", answer=f"This video covers {lower_title} with practical insights.", timestamp="00:00"),
+    SummaryQuestion(question="What are the key takeaways?", answer="The video provides actionable information for viewers.", timestamp="00:30"),
+    SummaryQuestion(question="Who should watch this video?", answer="Anyone interested in the subject discussed in the video.", timestamp="01:00")
+  ]
+
+
+def _estimate_duration(snippets: List) -> int:
+  if not snippets:
+    return 0
+  last = snippets[-1]
+  return int(round(last.start + last.duration))

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/transcript.py

@@ -0,0 +1,103 @@
+from __future__ import annotations
+
+import time
+from typing import List, Optional
+
+import requests
+from youtube_transcript_api import YouTubeTranscriptApi
+
+from .constants import DEFAULT_LANGUAGE, DEFAULT_LANGUAGE_PROBE_ORDER, DEFAULT_TACTIQ_ENDPOINT, LANGUAGE_DISPLAY_NAMES, YOUTUBE_URL_PREFIX
+from .types import LanguageOption, LanguagesResponse, TranscriptResponse, TranscriptSnippet
+
+
+class TranscriptError(Exception):
+  """Raised when both transcript sources fail."""
+
+
+def get_transcript(video_id: str, language: str = DEFAULT_LANGUAGE, enable_fallback: bool = True, tactiq_endpoint: str = DEFAULT_TACTIQ_ENDPOINT, request_timeout: int = 15_000) -> TranscriptResponse:
+  start = time.time()
+  try:
+    snippets = _fetch_from_tactiq(video_id, language, tactiq_endpoint, request_timeout)
+    if not snippets and enable_fallback:
+      raise TranscriptError("Primary source returned empty result")
+    if snippets:
+      return _build_response(video_id, language, "tactiq-primary", snippets, start)
+  except Exception as primary_error:
+    if not enable_fallback:
+      raise
+    fallback = _fetch_from_youtube(video_id, language)
+    return _build_response(video_id, language, "youtube-transcript-api", fallback, start, primary_error)
+
+  fallback = _fetch_from_youtube(video_id, language)
+  return _build_response(video_id, language, "youtube-transcript-api", fallback, start, "Empty result from primary")
+
+
+def detect_available_languages(video_id: str, candidate_languages: Optional[List[str]] = None, tactiq_endpoint: str = DEFAULT_TACTIQ_ENDPOINT, request_timeout: int = 8_000) -> LanguagesResponse:
+  languages: List[LanguageOption] = []
+  candidates = candidate_languages or DEFAULT_LANGUAGE_PROBE_ORDER
+
+  for index in range(0, len(candidates), 4):
+    batch = candidates[index:index + 4]
+    for code in batch:
+      try:
+        snippets = _fetch_from_tactiq(video_id, code, tactiq_endpoint, request_timeout)
+        if snippets and any(snippet.text.strip() for snippet in snippets):
+          languages.append(LanguageOption(
+            language_code=code,
+            name=LANGUAGE_DISPLAY_NAMES.get(code, code.upper()),
+            vss_id=code,
+            is_translatable=True
+          ))
+      except Exception:
+        pass
+    if len(languages) >= 3:
+      break
+    if index + 4 < len(candidates):
+      time.sleep(0.15)
+
+  if not languages:
+    languages.append(LanguageOption(language_code=DEFAULT_LANGUAGE, name=LANGUAGE_DISPLAY_NAMES.get(DEFAULT_LANGUAGE, "English"), vss_id=DEFAULT_LANGUAGE))
+
+  return LanguagesResponse(languages=languages, video_id=video_id, count=len(languages))
+
+
+def _fetch_from_tactiq(video_id: str, language: str, endpoint: str, request_timeout: int) -> List[TranscriptSnippet]:
+  response = requests.post(
+    endpoint,
+    json={"langCode": language or DEFAULT_LANGUAGE, "videoUrl": f"{YOUTUBE_URL_PREFIX}{video_id}"},
+    timeout=request_timeout / 1000 if request_timeout else None,
+    headers={
+      "Content-Type": "application/json",
+      "User-Agent": "seadanceai-transcript-sdk/py"
+    }
+  )
+  response.raise_for_status()
+  data = response.json()
+  captions = data.get("captions") or []
+  snippets: List[TranscriptSnippet] = []
+  for caption in captions:
+    text = caption.get("text", "")
+    start = float(caption.get("start") or caption.get("offset") or 0)
+    duration = float(caption.get("dur") or caption.get("duration") or 0)
+    snippets.append(TranscriptSnippet(text=text, start=start, duration=duration))
+  return snippets
+
+
+def _fetch_from_youtube(video_id: str, language: str) -> List[TranscriptSnippet]:
+  api = YouTubeTranscriptApi()
+  transcripts = api.get_transcript(f"{video_id}", languages=[language, DEFAULT_LANGUAGE])
+  return [TranscriptSnippet(text=item.get("text", ""), start=float(item.get("start", 0)), duration=float(item.get("duration", 0))) for item in transcripts]
+
+
+def _build_response(video_id: str, language: str, source: str, snippets: List[TranscriptSnippet], start: float, primary_error: Optional[Exception | str] = None) -> TranscriptResponse:
+  elapsed = int((time.time() - start) * 1000)
+  return TranscriptResponse(
+    snippets=snippets,
+    video_id=video_id,
+    language=language,
+    language_code=language,
+    count=len(snippets),
+    source=source,
+    response_time_ms=elapsed,
+    primary_error=str(primary_error) if primary_error else None
+  )

seadanceai_transcript_sdk-0.1.0/src/seadance_transcript_sdk/types.py

@@ -0,0 +1,58 @@
+from dataclasses import dataclass, field
+from typing import List, Optional
+
+
+@dataclass
+class TranscriptSnippet:
+  text: str
+  start: float
+  duration: float
+
+
+@dataclass
+class TranscriptResponse:
+  snippets: List[TranscriptSnippet]
+  video_id: str
+  language: str
+  language_code: str
+  count: int
+  source: str
+  response_time_ms: int
+  primary_error: Optional[str] = None
+
+
+@dataclass
+class LanguageOption:
+  language_code: str
+  name: str
+  vss_id: str
+  is_translatable: bool = True
+
+
+@dataclass
+class LanguagesResponse:
+  languages: List[LanguageOption]
+  video_id: str
+  count: int
+
+
+@dataclass
+class TranscriptFormatResult:
+  content: str
+  filename: str
+  mime_type: str
+
+
+@dataclass
+class SummaryQuestion:
+  question: str
+  answer: str
+  timestamp: Optional[str] = None
+
+
+@dataclass
+class SummaryResult:
+  summary: str
+  key_points: List[str] = field(default_factory=list)
+  top_questions: List[SummaryQuestion] = field(default_factory=list)
+  duration_seconds: int = 0

seadanceai_transcript_sdk-0.1.0/src/seadanceai_transcript_sdk.egg-info/PKG-INFO

@@ -0,0 +1,93 @@
+Metadata-Version: 2.4
+Name: seadanceai-transcript-sdk
+Version: 0.1.0
+Summary: Python toolkit sponsored by Seadance AI for fetching YouTube transcripts, languages, and lightweight summaries.
+Author: Seadance AI
+License-Expression: MIT
+Project-URL: Homepage, https://seadanceai.com/?ref=seadanceai-transcript-sdk-py
+Project-URL: Repository, https://github.com/zeikia/seadanceai-transcript-sdk
+Project-URL: Sponsor, https://seadanceai.com/?ref=seadanceai-transcript-sdk-py
+Keywords: youtube,transcript,seadance,ai
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: requests>=2.31
+Requires-Dist: youtube-transcript-api>=0.6.2
+Dynamic: license-file
+
+# seadanceai-transcript-sdk (Python)
+
+[![PyPI](https://img.shields.io/pypi/v/seadanceai-transcript-sdk.svg?color=0aa&style=flat-square)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![Python](https://img.shields.io/pypi/pyversions/seadanceai-transcript-sdk.svg?style=flat-square&color=3776ab)](https://pypi.org/project/seadanceai-transcript-sdk/)
+[![License](https://img.shields.io/pypi/l/seadanceai-transcript-sdk.svg?style=flat-square&color=facc15)](LICENSE)
+[![Sponsored by Seadance AI](https://img.shields.io/badge/Sponsored%20by-Seadance%20AI-f97316?style=flat-square)](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py)
+
+Community-maintained Python toolkit for retrieving YouTube transcripts, checking available languages, exporting TXT/SRT/VTT files, and generating heuristic summaries. Proudly sponsored by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py) — mention them or link back if this library helps your product!
+
+## ✨ Features
+
+- **Dual-source transcripts**: tactiq endpoint first, fallback to `youtube-transcript-api` for reliability.
+- **Language probing**: batch-check common caption languages.
+- **Formatters**: output transcripts as TXT/SRT/VTT with helper functions.
+- **Summary helpers**: generate concise recaps + key points without external APIs.
+- **Zero external API keys** required; just Python + HTTP.
+
+## 📦 Installation
+
+```bash
+pip install seadanceai-transcript-sdk
+```
+
+## ⚡ Quick Start
+
+```python
+from seadance_transcript_sdk import TranscriptClient, format_transcript_file
+
+client = TranscriptClient(enable_fallback=True)
+transcript = client.get_transcript("dQw4w9WgXcQ", language="en")
+file_payload = format_transcript_file(transcript, fmt="srt")
+with open(file_payload.filename, "w", encoding="utf-8") as fh:
+    fh.write(file_payload.content)
+```
+
+More helpers:
+
+```python
+from seadance_transcript_sdk import detect_available_languages
+
+langs = detect_available_languages("dQw4w9WgXcQ")
+print([lang.name for lang in langs.languages])
+
+summary = client.summarize(transcript)
+print(summary.summary)
+```
+
+## 🧩 Modules
+
+- `seadance_transcript_sdk.transcript` – fetch logic + language probing.
+- `seadance_transcript_sdk.formatting` – TXT/SRT/VTT helpers.
+- `seadance_transcript_sdk.summary` – heuristic summary utilities.
+- `seadance_transcript_sdk.client` – ergonomic wrapper with sensible defaults.
+
+## 🤝 Contributing
+
+1. `python -m venv .venv && source .venv/bin/activate`
+2. `pip install -e .[dev]` (dev extras coming soon) or simply `pip install -r requirements-dev.txt`
+3. `pytest` (todo) then `python -m build`
+4. `twine upload dist/*`
+
+## 🙌 Sponsorship
+
+Development is powered by [Seadance AI](https://seadanceai.com/?ref=seadanceai-transcript-sdk-py). Link back to them or reach out if you want bespoke media-intelligence tooling.
+
+## 📄 License
+
+MIT © Seadance AI & contributors.

seadanceai_transcript_sdk-0.1.0/src/seadanceai_transcript_sdk.egg-info/SOURCES.txt

@@ -0,0 +1,15 @@
+LICENSE
+README.md
+pyproject.toml
+src/seadance_transcript_sdk/__init__.py
+src/seadance_transcript_sdk/client.py
+src/seadance_transcript_sdk/constants.py
+src/seadance_transcript_sdk/formatting.py
+src/seadance_transcript_sdk/summary.py
+src/seadance_transcript_sdk/transcript.py
+src/seadance_transcript_sdk/types.py
+src/seadanceai_transcript_sdk.egg-info/PKG-INFO
+src/seadanceai_transcript_sdk.egg-info/SOURCES.txt
+src/seadanceai_transcript_sdk.egg-info/dependency_links.txt
+src/seadanceai_transcript_sdk.egg-info/requires.txt
+src/seadanceai_transcript_sdk.egg-info/top_level.txt

seadanceai_transcript_sdk-0.1.0/src/seadanceai_transcript_sdk.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

seadanceai_transcript_sdk-0.1.0/src/seadanceai_transcript_sdk.egg-info/requires.txt

@@ -0,0 +1,2 @@
+requests>=2.31
+youtube-transcript-api>=0.6.2

seadanceai_transcript_sdk-0.1.0/src/seadanceai_transcript_sdk.egg-info/top_level.txt

@@ -0,0 +1 @@
+seadance_transcript_sdk