Audio Not Generating from Podcast Transcript despite format

[e-ratsavong]

Hi
When I try to generate audio, my mp3 files are all empty. This is the code I'm using to generate my audio files:

audio_file_from_transcript = generate_podcast(
    transcript_file="./data/transcripts/transcript_test.txt",
    is_local=True,
    tts_model="edge"
)

my transcript was at first this:

<Person1> "Welcome back to PODCASTIFY! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends."
</Person1><Person2>  "This is exciting. I'm ready to learn more about what's going on in the global solar energy sector!"  
</Person2><Person1> "Excellent! So, according to this article from pv magazine International, we're seeing a rebound in solar wafer prices. However, it's not quite reaching its limit. Let's break down the details." 
</Person1>
...

I wasn't sure if the quotations were creating errors with generating the audio so I changed it to this:

<Person1> Welcome back to PODCASTIFY! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends. </Person1>
<Person2> This is exciting. I'm ready to learn more about what's going on in the global solar energy sector! </Person2>

But I am still getting the same issue. There are no console errors just blank audio files that are 428 bytes.

This is the output name of mp3
podcast_a401a2fdc500484380fca7593bb34cae.mp3

[souzatharsis]

Hi, Please try without the param 'is_local'. Also, make sure ffmpeg and its dependencies are properly installed. Best, Tharsis. <http://linkedin.com/in/tharsissouza>

…

On Mon, Feb 17, 2025 at 2:17 PM e-ratsavong ***@***.***> wrote: Hi When I try to generate audio, my mp3 files are all empty. This is the code I'm using to generate my audio files: audio_file_from_transcript = generate_podcast( transcript_file="./data/transcripts/transcript_test.txt", is_local=True, tts_model="edge" ) my transcript was at first this: <Person1> "Welcome back to PODCASTIFY! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends." </Person1><Person2> "This is exciting. I'm ready to learn more about what's going on in the global solar energy sector!" </Person2><Person1> "Excellent! So, according to this article from pv magazine International, we're seeing a rebound in solar wafer prices. However, it's not quite reaching its limit. Let's break down the details." </Person1> ... I wasn't sure if the quotations were creating errors with generating the audio so I changed it to this: <Person1> Welcome back to Renewables Energy News Flash! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends. </Person1> <Person2> This is exciting. I'm ready to learn more about what's going on in the global solar energy sector! </Person2> But I am still getting the same issue. There are no console errors just blank audio files that are 428 bytes. This is the output name of mp3 podcast_a401a2fdc500484380fca7593bb34cae.mp3 — Reply to this email directly, view it on GitHub <#243>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3PEQG4BK2XFXRZRCMD2QIYSTAVCNFSM6AAAAABXKAMXNSVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA2TQNRSHE3DCMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> [image: e-ratsavong]*e-ratsavong* created an issue (souzatharsis/podcastfy#243) <#243> Hi When I try to generate audio, my mp3 files are all empty. This is the code I'm using to generate my audio files: audio_file_from_transcript = generate_podcast( transcript_file="./data/transcripts/transcript_test.txt", is_local=True, tts_model="edge" ) my transcript was at first this: <Person1> "Welcome back to PODCASTIFY! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends." </Person1><Person2> "This is exciting. I'm ready to learn more about what's going on in the global solar energy sector!" </Person2><Person1> "Excellent! So, according to this article from pv magazine International, we're seeing a rebound in solar wafer prices. However, it's not quite reaching its limit. Let's break down the details." </Person1> ... I wasn't sure if the quotations were creating errors with generating the audio so I changed it to this: <Person1> Welcome back to Renewables Energy News Flash! Today, we're diving deep into the solar wafer market with a fascinating analysis of prices and trends. </Person1> <Person2> This is exciting. I'm ready to learn more about what's going on in the global solar energy sector! </Person2> But I am still getting the same issue. There are no console errors just blank audio files that are 428 bytes. This is the output name of mp3 podcast_a401a2fdc500484380fca7593bb34cae.mp3 — Reply to this email directly, view it on GitHub <#243>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3PEQG4BK2XFXRZRCMD2QIYSTAVCNFSM6AAAAABXKAMXNSVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA2TQNRSHE3DCMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[e-ratsavong]

@souzatharsis I removed the is_local command and tried with other tts model besides edge. I also ran ffmpeg manually/separately and was able to generate an audio file. But when I try to generate the audio using podcastfy and try to open the audio file it says "We can't open [file name] this may be because the file type is unsupported, the file extension is incorrect, or the file is corrupt". When I inspect the actual .mp3 it says: Error occurred: [mp3 @ 000002546984CB80] Failed to find two consecutive MPEG audio frames.
data/audio/podcast_70eee2bca48f4527a5fc8a73062a4ec9.mp3: Invalid data found when processing input

[souzatharsis]

Windows OS? <http://linkedin.com/in/tharsissouza>

…

On Mon, Feb 17, 2025 at 3:27 PM e-ratsavong ***@***.***> wrote: @souzatharsis <https://github.com/souzatharsis> I removed the is_local command and tried with other tts model besides edge. I also ran ffmpeg manually/separately and was able to generate an audio file. But when I try to generate the audio using podcastfy and try to open the audio file it says "We can't open [file name] this may be because the file type is unsupported, the file extension is incorrect, or the file is corrupt". When I inspect the actual .mp3 it says: Error occurred: [mp3 @ 000002546984CB80] Failed to find two consecutive MPEG audio frames. data/audio/podcast_70eee2bca48f4527a5fc8a73062a4ec9.mp3: Invalid data found when processing input — Reply to this email directly, view it on GitHub <#243 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3JSF52BFQ76SU2DNL32QJA4DAVCNFSM6AAAAABXKAMXNSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNRUGAYTOMZYHA> . You are receiving this because you were mentioned.Message ID: ***@***.***> [image: e-ratsavong]*e-ratsavong* left a comment (souzatharsis/podcastfy#243) <#243 (comment)> @souzatharsis <https://github.com/souzatharsis> I removed the is_local command and tried with other tts model besides edge. I also ran ffmpeg manually/separately and was able to generate an audio file. But when I try to generate the audio using podcastfy and try to open the audio file it says "We can't open [file name] this may be because the file type is unsupported, the file extension is incorrect, or the file is corrupt". When I inspect the actual .mp3 it says: Error occurred: [mp3 @ 000002546984CB80] Failed to find two consecutive MPEG audio frames. data/audio/podcast_70eee2bca48f4527a5fc8a73062a4ec9.mp3: Invalid data found when processing input — Reply to this email directly, view it on GitHub <#243 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADTMY3JSF52BFQ76SU2DNL32QJA4DAVCNFSM6AAAAABXKAMXNSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNRUGAYTOMZYHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[e-ratsavong]

Yes