Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
8 changes: 7 additions & 1 deletion src/khoj/processor/speech/text_to_speech.py
Original file line number Diff line number Diff line change
Expand Up @@ -17,12 +17,18 @@ def is_eleven_labs_enabled():
return ELEVEN_LABS_API_KEY is not None


class TextToSpeechError(Exception):
"""Exception raised when text-to-speech generation fails."""

pass


def generate_text_to_speech(
text_to_speak: str,
voice_id: str = VOICE_ID,
):
if not is_eleven_labs_enabled():
return "Eleven Labs API key is not set"
raise TextToSpeechError("Eleven Labs API key is not set")

# Convert the incoming text from markdown format to plain text
html = markdown_renderer.render(text_to_speak)
Expand Down
7 changes: 5 additions & 2 deletions src/khoj/routers/api_chat.py
Original file line number Diff line number Diff line change
Expand Up @@ -43,7 +43,7 @@
)
from khoj.processor.image.generate import text_to_image
from khoj.processor.operator import operate_environment
from khoj.processor.speech.text_to_speech import generate_text_to_speech
from khoj.processor.speech.text_to_speech import TextToSpeechError, generate_text_to_speech
from khoj.processor.tools.online_search import (
deduplicate_organic_results,
read_webpages,
Expand Down Expand Up @@ -208,7 +208,10 @@ async def text_to_speech(
if voice_model:
params["voice_id"] = voice_model.model_id

speech_stream = generate_text_to_speech(**params)
try:
speech_stream = generate_text_to_speech(**params)
except TextToSpeechError as e:
raise HTTPException(status_code=501, detail=str(e))
return StreamingResponse(speech_stream.iter_content(chunk_size=1024), media_type="audio/mpeg")


Expand Down