Skip to content

feat: Add Japanese language support for podcast generation #11

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat: Add Japanese language support for podcast generation #11

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 19 additions & 7 deletions backend/app/routers/generate.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,10 @@
from app.core.limiter import limiter
import os
from app.prompts import PODCAST_SSML_PROMPT_AFTER_BREAK, PODCAST_SSML_PROMPT, PODCAST_SSML_PROMPT_BEFORE_BREAK, SLIDE_PROMPT
from app.prompts_jp import PODCAST_SSML_PROMPT_AFTER_BREAK as PODCAST_SSML_PROMPT_AFTER_BREAK_JP
from app.prompts_jp import PODCAST_SSML_PROMPT as PODCAST_SSML_PROMPT_JP
from app.prompts_jp import PODCAST_SSML_PROMPT_BEFORE_BREAK as PODCAST_SSML_PROMPT_BEFORE_BREAK_JP
from app.prompts_jp import SLIDE_PROMPT as SLIDE_PROMPT_JP
from anthropic._exceptions import RateLimitError
from pydantic import BaseModel
from functools import lru_cache
Expand Down Expand Up @@ -125,11 +129,12 @@ def process_github_content_for_slides(content, slide_prompt, max_length, max_tok
return slide_markdown_response


def generate_ssml_concurrently(file_tree, readme, file_content, audio_length) -> str | dict:
def generate_ssml_concurrently(file_tree, readme, file_content, audio_length, language='en') -> str | dict:
# Prepare the content
if audio_length == 'short':
combined_content = f"FILE TREE: {file_tree}\nREADME: {readme} IMPORTANT FILES: {file_content}"
ssml_response = process_github_content(combined_content, PODCAST_SSML_PROMPT, 250000, 100000)
prompt = PODCAST_SSML_PROMPT_JP if language == 'ja' else PODCAST_SSML_PROMPT
ssml_response = process_github_content(combined_content, prompt, 250000, 100000)
Comment on lines +136 to +137
Copy link

@gemini-code-assist gemini-code-assist bot Jul 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The prompt selection logic is duplicated. Consider centralizing the prompt configuration to avoid repetition and improve maintainability. This applies to lines 151, 183, and 292 as well.

return ssml_response
else:
combined_content_tree_readme = f"FILE TREE: {file_tree}\nREADME: {readme}"
Expand All @@ -143,17 +148,20 @@ def check_response(response):

# Use ThreadPoolExecutor to execute tasks concurrently
with concurrent.futures.ThreadPoolExecutor() as executor:
prompt_before = PODCAST_SSML_PROMPT_BEFORE_BREAK_JP if language == 'ja' else PODCAST_SSML_PROMPT_BEFORE_BREAK
prompt_after = PODCAST_SSML_PROMPT_AFTER_BREAK_JP if language == 'ja' else PODCAST_SSML_PROMPT_AFTER_BREAK

future_tree_readme = executor.submit(
process_github_content,
combined_content_tree_readme,
PODCAST_SSML_PROMPT_BEFORE_BREAK,
prompt_before,
250000,
100000
)
future_file_content = executor.submit(
process_github_content,
combined_content_file_content,
PODCAST_SSML_PROMPT_AFTER_BREAK,
prompt_after,
250000,
100000
)
Expand All @@ -172,7 +180,8 @@ def check_response(response):
combined_ssml_content = f"{ssml_response_tree_readme_content}\n{ssml_response_file_content_content}"

# Wrap the combined content in a single <speak> tag
full_ssml_response = f'<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="en-US">{combined_ssml_content}</speak>'
xml_lang = 'ja-JP' if language == 'ja' else 'en-US'
full_ssml_response = f'<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{xml_lang}">{combined_ssml_content}</speak>'

# Proceed with ssml_response_tree_readme and ssml_response_file_content as needed
return full_ssml_response
Expand All @@ -185,6 +194,7 @@ class ApiRequest(BaseModel):
api_key: str | None = None
audio: bool = False # new param
audio_length: str = 'long'
language: str = 'en' # new param for language selection
Copy link

@gemini-code-assist gemini-code-assist bot Jul 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

Consider using typing.Literal to restrict the language field to supported values ('en', 'ja') for improved data validation. This should also be applied to the SlideRequest model on line 207.

Suggested change
language: str = 'en' # new param for language selection
language: Literal['en', 'ja'] = 'en' # new param for language selection



class SlideRequest(BaseModel):
Expand All @@ -194,6 +204,7 @@ class SlideRequest(BaseModel):
api_key: str | None = None
audio: bool = False # new param
audio_length: str = 'long'
language: str = 'en' # new param for language selection


# @limiter.limit("1/minute;5/day") # TEMP: disable rate limit for growth??
Expand All @@ -216,7 +227,7 @@ async def generate(request: Request, body: ApiRequest):
readme = github_data["readme"]
file_content = github_data["file_content"]

result = generate_ssml_concurrently(file_tree, readme, file_content, audio_length)
result = generate_ssml_concurrently(file_tree, readme, file_content, audio_length, body.language)
# Check if there was an error response
if isinstance(result, dict): # There was an error
print("Error in processing:")
Expand Down Expand Up @@ -278,7 +289,8 @@ async def generate_slide(request: Request, body: SlideRequest):
file_tree = github_data["file_tree"]
readme = github_data["readme"]
file_content = github_data["file_content"]
markdown = process_github_content_for_slides(f" file tree: {file_tree} \n contents: {file_content}", SLIDE_PROMPT, 250000, 100000)
slide_prompt = SLIDE_PROMPT_JP if body.language == 'ja' else SLIDE_PROMPT
markdown = process_github_content_for_slides(f" file tree: {file_tree} \n contents: {file_content}", slide_prompt, 250000, 100000)
return {"slide_markdown": markdown}
except Exception as e:
raise HTTPException(status_code=500, detail=f"An error occurred: {str(e)}")
Expand Down
Loading