Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
ralyodio closed an issue on souzatharsis/podcastfy
suddenly getting error
``` Error generating content: "Input to ChatPromptTemplate is missing variables {'input_texts'}. Expected: ['conversation_style', 'dialogue_structure', 'engagement_techniques', 'input_text', 'inp...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
pip install --upgrade podcastfy On Fri, Oct 25, 2024, 9:50 AM ralyodio ***@***.***> wrote: > how do i update? > > — > Reply to this email directly, view it on GitHub > <https://github.com...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
Please update to latest Let me know if error persists On Fri, Oct 25, 2024, 9:44 AM ralyodio ***@***.***> wrote: > Error generating content: "Input to ChatPromptTemplate is missing variabl...
ralyodio opened an issue on souzatharsis/podcastfy
suddenly getting error
``` Error generating content: "Input to ChatPromptTemplate is missing variables {'input_texts'}. Expected: ['conversation_style', 'dialogue_structure', 'engagement_techniques', 'input_text', 'inp...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
That's good feedback. I'll make sure all markup is dropped. On Fri, Oct 25, 2024, 8:20 AM Kuniaki Shimizu ***@***.***> wrote: > That is, is the speaker actually saying a tag out loud? > > ...
kun432 created a comment on an issue on souzatharsis/podcastfy
>That is, is the speaker actually saying a tag out loud? yep, it sounds like "speak (japanese speech continues)". This happens not always like it sometime is spoken, usually ignored, depends on ...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
But is it compromising audio generation in your experience? That is, is the speaker actually saying a tag out loud? Anyhow that's a good point, I've been thinking about dropping all TTS markup...
kun432 created a comment on an issue on souzatharsis/podcastfy
I mean no SSML tag should be passed to TTS generation especially for OpenAI because OpenAI TTS does not support SSML, I think. So `clean_tss_markup` should remove all SSML tag for OpenAI TTS, but i...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
Hi, the SSML markup is generated by llm hence not guaranteed to always be generated. What are you trying to achieve with the speak tag? On Fri, Oct 25, 2024, 5:16 AM Kuniaki Shimizu ***@***....
kun432 created a comment on an issue on souzatharsis/podcastfy
this does not always happen. Depending on transcripts generates by LLM, sometimes it includes "<speak>", sometimes not.
kun432 opened an issue on souzatharsis/podcastfy
[Question] does OpenAI TTS supports SSML?
I generated audio with a pdf and custom config with Japanese and changing default TTS model like this: ```python from podcastfy.client import generate_podcast audio_from_local_pdf = generate...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
Another advantage besides open and free of xtts_v2 is that you can enable voice cloning. https://huggingface.co/coqui/XTTS-v2 ``` python from TTS.api import TTS tts = TTS("tts_models/multil...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
simply adding os.environ["LANGCHAIN_TRACING_V2"] = "false" should remove the warning I'll push an update with it by default
souzatharsis created a comment on a pull request on souzatharsis/podcastfy
automated tests indicate api keys are not found. please check above logs.
souzatharsis closed an issue on souzatharsis/podcastfy
Please prune the dependencies in requirements.txt
Hi @souzatharsis, thanks for making this open source. I don't think all the packages listed in `requirements.txt` are direct dependencies. You can list only the primary dependencies while removi...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
Do you still have this problem. If not, please close the issue.
souzatharsis created a comment on an issue on souzatharsis/podcastfy
please try again. Audio from pdf had a bug that has been fixed now.
souzatharsis closed an issue on souzatharsis/podcastfy
Audio samples for free
The link to the audio files requires a subscription. Are they located anywhere else
souzatharsis closed an issue on souzatharsis/podcastfy
how can I brand the audio output?
I have a podcast called "Byte Broadcast" -- can I make it reference that name?
souzatharsis created a comment on an issue on souzatharsis/podcastfy
This is out-of-scope. Please use any off-the-shelf audio editing tool instead.
souzatharsis closed an issue on souzatharsis/podcastfy
include bumper music
Can I add a fade in and fade out bumper audio clip (ie: soundtrack music of 10 seconds or something)?
souzatharsis created a tag on souzatharsis/podcastfy
v0.2.8 - An Open Source alternative to NotebookLM's podcast feature: Transforming Multimodal Content into Captivating Multilingual Audio Conversations with GenAI
souzatharsis published a release on souzatharsis/podcastfy
v0.2.8
## What's Changed * Feat/input raw text by @souzatharsis in https://github.com/souzatharsis/podcastfy/pull/108 * feat: enable users to customize conversation with free text. Also fix… by @souzath...
souzatharsis created a comment on an issue on souzatharsis/podcastfy
@CryptoCultCurt This has been fixed now. Please let me know if it now works for you.
souzatharsis closed an issue on souzatharsis/podcastfy
TTS param for custom conversation not working