Recherche avancée

Sur d’autres sites (10383)

• ffmpeg failed to load audio file

  14 avril 2024, par Vaishnav Ghenge

  Failed to load audio: ffmpeg version 5.1.4-0+deb12u1 Copyright (c) Failed to load audio: ffmpeg version 5.1.4-0+deb12u1 Copyright (c) 2000-2023 the FFmpeg developers
  built with gcc 12 (Debian 12.2.0-14)
  configuration: --prefix=/usr --extra-version=0+deb12u1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --arch=amd64 --enable-gpl --disable-stripping --enable-gnutls --enable-ladspa --enable-libaom --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libcodec2 --enable-libdav1d --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libglslang --enable-libgme --enable-libgsm --enable-libjack --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librabbitmq --enable-librist --enable-librubberband --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libsrt --enable-libssh --enable-libsvtav1 --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzimg --enable-libzmq --enable-libzvbi --enable-lv2 --enable-omx --enable-openal --enable-opencl --enable-opengl --enable-sdl2 --disable-sndio --enable-libjxl --enable-pocketsphinx --enable-librsvg --enable-libmfx --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-chromaprint --enable-frei0r --enable-libx264 --enable-libplacebo --enable-librav1e --enable-shared
  libavutil      57. 28.100 / 57. 28.100
  libavcodec     59. 37.100 / 59. 37.100
  libavformat    59. 27.100 / 59. 27.100
  libavdevice    59.  7.100 / 59.  7.100
  libavfilter     8. 44.100 /  8. 44.100
  libswscale      6.  7.100 /  6.  7.100
  libswresample   4.  7.100 /  4.  7.100
  libpostproc    56.  6.100 / 56.  6.100
/tmp/tmpjlchcpdm.wav: Invalid data found when processing input


  


  backend :

  


  
@app.route("/transcribe", methods=["POST"])
def transcribe():
    # Check if audio file is present in the request
    if 'audio_file' not in request.files:
        return jsonify({"error": "No file part"}), 400
    
    audio_file = request.files.get('audio_file')

    # Check if audio_file is sent in files
    if not audio_file:
        return jsonify({"error": "`audio_file` is missing in request.files"}), 400

    # Check if the file is present
    if audio_file.filename == '':
        return jsonify({"error": "No selected file"}), 400

    # Save the file with a unique name
    filename = secure_filename(audio_file.filename)
    unique_filename = os.path.join("uploads", str(uuid.uuid4()) + '_' + filename)
    # audio_file.save(unique_filename)
    
    # Read the contents of the audio file
    contents = audio_file.read()

    max_file_size = 500 * 1024 * 1024
    if len(contents) > max_file_size:
        return jsonify({"error": "File is too large"}), 400

    # Check if the file extension suggests it's a WAV file
    if not filename.lower().endswith('.wav'):
        # Delete the file if it's not a WAV file
        os.remove(unique_filename)
        return jsonify({"error": "Only WAV files are supported"}), 400

    print(f"\033[92m{filename}\033[0m")

    # Call Celery task asynchronously
    result = transcribe_audio.delay(contents)

    return jsonify({
        "task_id": result.id,
        "status": "pending"
    })


@celery_app.task
def transcribe_audio(contents):
    # Transcribe the audio
    try:
        # Create a temporary file to save the audio data
        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as temp_audio:
            temp_path = temp_audio.name
            temp_audio.write(contents)

            print(f"\033[92mFile temporary path: {temp_path}\033[0m")
            transcribe_start_time = time.time()

            # Transcribe the audio
            transcription = transcribe_with_whisper(temp_path)
            
            transcribe_end_time = time.time()
            print(f"\033[92mTranscripted text: {transcription}\033[0m")

            return transcription, transcribe_end_time - transcribe_start_time

    except Exception as e:
        print(f"\033[92mError: {e}\033[0m")
        return str(e)


  


  frontend :

  


      useEffect(() => {
        const init = () => {
            navigator.mediaDevices.getUserMedia({audio: true})
                .then((audioStream) => {
                    const recorder = new MediaRecorder(audioStream);

                    recorder.ondataavailable = e => {
                        if (e.data.size > 0) {
                            setChunks(prevChunks => [...prevChunks, e.data]);
                        }
                    };

                    recorder.onerror = (e) => {
                        console.log("error: ", e);
                    }

                    recorder.onstart = () => {
                        console.log("started");
                    }

                    recorder.start();

                    setStream(audioStream);
                    setRecorder(recorder);
                });
        }

        init();

        return () => {
            if (recorder && recorder.state === 'recording') {
                recorder.stop();
            }

            if (stream) {
                stream.getTracks().forEach(track => track.stop());
            }
        }
    }, []);

    useEffect(() => {
        // Send chunks of audio data to the backend at regular intervals
        const intervalId = setInterval(() => {
            if (recorder && recorder.state === 'recording') {
                recorder.requestData(); // Trigger data available event
            }
        }, 8000); // Adjust the interval as needed


        return () => {
            if (intervalId) {
                console.log("Interval cleared");
                clearInterval(intervalId);
            }
        };
    }, [recorder]);

    useEffect(() => {
        const processAudio = async () => {
            if (chunks.length > 0) {
                // Send the latest chunk to the server for transcription
                const latestChunk = chunks[chunks.length - 1];

                const audioBlob = new Blob([latestChunk]);
                convertBlobToAudioFile(audioBlob);
            }
        };

        void processAudio();
    }, [chunks]);

    const convertBlobToAudioFile = useCallback((blob: Blob) => {
        // Convert Blob to audio file (e.g., WAV)
        // This conversion may require using a third-party library or service
        // For example, you can use the MediaRecorder API to record audio in WAV format directly
        // Alternatively, you can use a library like recorderjs to perform the conversion
        // Here's a simplified example using recorderjs:

        const reader = new FileReader();
        reader.onload = () => {
            const audioBuffer = reader.result; // ArrayBuffer containing audio data

            // Send audioBuffer to Flask server or perform further processing
            sendAudioToFlask(audioBuffer as ArrayBuffer);
        };

        reader.readAsArrayBuffer(blob);
    }, []);

    const sendAudioToFlask = useCallback((audioBuffer: ArrayBuffer) => {
        const formData = new FormData();
        formData.append('audio_file', new Blob([audioBuffer]), `speech_audio.wav`);

        console.log(formData.get("audio_file"));

        fetch('http://34.87.75.138:8000/transcribe', {
            method: 'POST',
            body: formData
        })
            .then(response => response.json())
            .then((data: { task_id: string, status: string }) => {
                pendingTaskIdsRef.current.push(data.task_id);
            })
            .catch(error => {
                console.error('Error sending audio to Flask server:', error);
            });
    }, []);


  


  I was trying to pass the audio from frontend to whisper model which is in flask app

  


• VLC Player shows broken HLS stream with 4k HDR10 mkv

  8 avril 2023, par goodkid38

  I am trying to convert a 4k mkv to an HLS stream but I am not having any luck. I have tried a few ffmpeg commands to try and fix the issue but none have worked. Here are the commands I have tried.

  


  

  1. Basic copy command :

  


  


  ffmpeg -i "video.mkv" -c copy -f hls "plexTemp/out.m3u8"

  


  

  2. Command used to see if it was an HDR color issue :

  


  


  ffmpeg -i "video.mkv" -c copy -pix_fmt yuv420p10le -f hls "plexTemp/out.m3u8"
3. Command used to revert to 8 bit color :

  


  ffmpeg -i "video.mkv" -c copy -pix_fmt yuv420p -f hls "plexTemp/out.m3u8"

  


  

  4. I tried removing extra streams and just focusing on audio and video :

  


  


  ffmpeg -i "out.mkv" -map 0:v:0 -map 0:a:1 -c copy -pix_fmt yuv420p10le -f hls "plexTemp/out.m3u8"

  


  I also saw these warnings when running each command.Stream HEVC is not hvc1, you should use tag:v hvc1 to set it.
And this

  


  [matroska,webm @ 000001d7921803c0] Stream #12: not enough frames to estimate rate; consider increasing probesize
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 6 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 7 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 8 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 9 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 10 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options
[matroska,webm @ 000001d7921803c0] Could not find codec parameters for stream 11 (Subtitle: hdmv_pgs_subtitle (pgssub)): unspecified size
Consider increasing the value for the 'analyzeduration' (0) and 'probesize' (5000000) options


  


  So I tried increasing the analyzeduration and probesize and adding the tag like so :
ffmpeg -analyzeduration 10000000 -probesize 10000000 -i "out.mkv" -c copy -tag:v hvc1 -f hls "plexTemp/out.m3u8"

  


  To no avail. Here is what the output looks like on VLC. It's mostly black with a few lines of color that randomly change.

  


  On my TV I see this :

  


  

  


  One thing I see that stands out in ffmpegs output is this :

  


  [hls @ 00000207239a9ec0] Opening 'plexTemp/out0.ts' for writing7 bitrate=  -0.0kbits/s speed=N/A
[hls @ 00000207239a9ec0] Opening 'plexTemp/out.m3u8.tmp' for writing


  


  Here is the audio and video info on the mkv :

  


  General
Unique ID                                : 92280908398971492516286250889389584022 (0x456CA80EF29B1357B572719D6EC4AE96)
Complete name                            : I:\video.mkv
Format                                   : Matroska
Format version                           : Version 2
File size                                : 49.4 GiB
Duration                                 : 1 h 39 min
Overall bit rate mode                    : Variable
Overall bit rate                         : 71.2 Mb/s
Frame rate                               : 23.976 FPS
Movie name                               : video
Encoded date                             : 2023-04-06 22:39:53 UTC
Writing application                      : MakeMKV v1.16.7 win(x64-release)
Writing library                          : libmakemkv v1.16.7 (1.3.10/1.5.2) win(x64-release)
Cover                                    : Yes
Attachments                              : cover.jpg

Video
ID                                       : 1
ID in the original source medium         : 4113 (0x1011)
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@High
HDR format                               : SMPTE ST 2086, HDR10 compatible
Codec ID                                 : V_MPEGH/ISO/HEVC
Duration                                 : 1 h 39 min
Bit rate                                 : 63.6 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 23.976 (24000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0 (Type 2)
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.320
Stream size                              : 44.1 GiB (89%)
Writing library                          : ATEME Titan File 3.9.6 (4.9.6.2)        
Language                                 : English
Default                                  : No
Forced                                   : No
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0050 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 1000 cd/m2
Maximum Frame-Average Light Level        : 140 cd/m2
Original source medium                   : Blu-ray

Audio #1
ID                                       : 2
ID in the original source medium         : 4352 (0x1100)
Format                                   : DTS XLL X
Format/Info                              : Digital Theater Systems
Commercial name                          : DTS:X
Codec ID                                 : A_DTS
Duration                                 : 1 h 39 min
Bit rate mode                            : Variable
Bit rate                                 : 4 174 kb/s
Channel(s)                               : 8 channels
Channel layout                           : C L R LFE Lb Rb Lss Rss
Sampling rate                            : 48.0 kHz
Frame rate                               : 93.750 FPS (512 SPF)
Bit depth                                : 24 bits
Stream size                              : 2.89 GiB (6%)
Title                                    : Surround 7.1
Language                                 : English
Default                                  : Yes
Forced                                   : No
Original source medium                   : Blu-ray
Here is information on the HLS output:
General
Complete name                            : I:\out.m3u8
Format                                   : HLS
Format profile                           : Media
File size                                : 67.4 MiB
Duration                                 : 8 s 138 ms
Overall bit rate mode                    : Variable
Overall bit rate                         : 69.4 Mb/s
Frame rate                               : 23.976 FPS


  


  And my
Here is the output of my HLS stream :

  


  Video
ID                                       : 256 (0x100)
Menu ID                                  : 1 (0x1)
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@High
HDR format                               : SMPTE ST 2086, HDR10 compatible
Muxing mode                              : MPEG-TS
Codec ID                                 : 36
Duration                                 : 8 s 49 ms
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 (24000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0 (Type 2)
Bit depth                                : 10 bits
Writing library                          : ATEME Titan File 3.9.6 (4.9.6.2)        
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0050 cd/m2, max: 1000 cd/m2
Maximum Content Light Level              : 1000 cd/m2
Maximum Frame-Average Light Level        : 140 cd/m2
Source                                   : out92.ts

Audio
ID                                       : 257 (0x101)
Menu ID                                  : 1 (0x1)
Format                                   : DTS XLL X
Format/Info                              : Digital Theater Systems
Commercial name                          : DTS:X
Muxing mode                              : MPEG-TS
Codec ID                                 : 130
Duration                                 : 8 s 138 ms
Bit rate mode                            : Variable
Channel(s)                               : 8 channels
Channel layout                           : C L R LFE Lb Rb Lss Rss
Sampling rate                            : 48.0 kHz
Frame rate                               : 93.750 FPS (512 SPF)
Bit depth                                : 24 bits
Delay relative to video                  : -125 ms
Language                                 : English
Source                                   : out92.ts


  


• A Guide to GDPR Sensitive Personal Data

  13 mai 2024, par Erin

  The General Data Protection Regulation (GDPR) is one of the world’s most stringent data protection laws. It provides a legal framework for collection and processing of the personal data of EU individuals.

  The GDPR distinguishes between “special categories of personal data” (also referred to as “sensitive”) and other personal data and imposes stricter requirements on collection and processing of sensitive data. Understanding these differences will help your company comply with the requirements and avoid heavy penalties.

  In this article, we’ll explain what personal data is considered “sensitive” according to the GDPR. We’ll also examine how a web analytics solution like Matomo can help you maintain compliance.

  What is sensitive personal data ?

  The following categories of data are treated as sensitive :

  1. 1. Personal data revealing :
        • Racial or ethnic origin ;
        • Political opinions ;
        • Religious or philosophical beliefs ;
        • Trade union membership ;
     2. Genetic and biometric data ;
     3. Data concerning a person’s :
        • Health ; or
        • Sex life or sexual orientation.

  

  Sensitive vs. non-sensitive personal data : What’s the difference ?

  While both categories include information about an individual, sensitive data is seen as more private, or requiring a greater protection. 

  Sensitive data often carries a higher degree of risk and harm to the data subject, if the data is exposed. For example, a data breach exposing health records could lead to discrimination for the individuals involved. An insurance company could use the information to increase premiums or deny coverage. 

  In contrast, personal data like name or gender is considered less sensitive because it doesn’t carry the same degree of harm as sensitive data. 

  Unauthorised access to someone’s name alone is less likely to harm them or infringe on their fundamental rights and freedoms than an unauthorised access to their health records or biometric data. Note that financial information (e.g. credit card details) does not fall into the special categories of data.

  

  Legality of processing

  Under the GDPR, both sensitive and nonsensitive personal data are protected. However, the rules and conditions for processing sensitive data are more stringent.

  Article 6 deals with processing of non-sensitive data and it states that processing is lawful if one of the six lawful bases for processing applies. 

  In contrast, Art. 9 of the GDPR states that processing of sensitive data is prohibited as a rule, but provides ten exceptions. 

  It is important to note that the lawful bases in Art. 6 are not the same as exceptions in Art. 9. For example, while performance of a contract or legitimate interest of the controller are a lawful basis for processing non-sensitive personal data, they are not included as an exception in Art. 9. What follows is that controllers are not permitted to process sensitive data on the basis of contract or legitimate interest. 

  The exceptions where processing of sensitive personal data is permitted (subject to additional requirements) are : 

  • Explicit consent : The individual has given explicit consent to processing their sensitive personal data for specified purpose(s), except where an EU member state prohibits such consent. See below for more information about explicit consent. 
  • Employment, social security or social protection : Processing sensitive data is necessary to perform tasks under employment, social security or social protection law.
  • Vital interests : Processing sensitive data is necessary to protect the interests of a data subject or if the individual is physically or legally incapable of consenting. 
  • Non-for-profit bodies : Foundations, associations or nonprofits with a political, philosophical, religious or trade union aim may process the sensitive data of their members or those they are in regular contact with, in connection with their purposes (and no disclosure of the data is permitted outside the organisation, without the data subject’s consent).
  • Made public : In some cases, it may be permissible to process the sensitive data of a data subject if the individual has already made it public and accessible. 
  • Legal claims : Processing sensitive data is necessary to establish, exercise or defend legal claims, including legal or in court proceedings.
  • Public interest : Processing is necessary for reasons of substantial public interest, like preventing unlawful acts or protecting the public.
  • Health or social care : Processing special category data is necessary for : preventative or occupational medicine, providing health and social care, medical diagnosis or managing healthcare systems.
  • Public health : It is permissible to process sensitive data for public health reasons, like protecting against cross-border threats to health or ensuring the safety of medicinal products or medical devices. 
  • Archiving, research and statistics : You may process sensitive data if it’s done for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes.

  In addition, you must adhere to all data handling requirements set by the GDPR.

  Important : Note that for any data sent that you are processing, you always need to identify a lawful basis under Art. 6. In addition, if the data sent contains sensitive data, you must comply with Art. 9.

  Explicit consent

  While consent is a valid lawful basis for processing non-sensitive personal data, controllers are permitted to process sensitive data only with an “explicit consent” of the data subject.

  The GDPR does not define “explicit” consent, but it is accepted that it must meet all Art. 7 conditions for consent, at a higher threshold. To be “explicit” a consent requires a clear statement (oral or written) of the data subject. Consent inferred from the data subject’s actions does not meet the threshold. 

  The controller must retain records of the explicit consent and provide appropriate consent withdrawal method to allow the data subject to exercise their rights.

  Examples of compliant and non-compliant sensitive data processing

  Here are examples of when you can and can’t process sensitive data :

  • When you can process sensitive data : A doctor logs sensitive data about a patient, including their name, symptoms and medicine prescribed. The hospital can process this data to provide appropriate medical care to their patients. An IoT device and software manufacturer processes their customers’ health data based on explicit consent of each customer. 
  • When you can’t process sensitive data : One example is when you don’t have explicit consent from a data subject. Another is when there’s no lawful basis for processing it or you are collecting personal data you simply do not need. For example, you don’t need your customer’s ethnic origin to fulfil an online order.

  Other implications of processing sensitive data

  If you process sensitive data, especially on a large scale, GDPR imposes additional requirements, such as having Data Privacy Impact Assessments, appointing Data Protection Officers and EU Representatives, if you are a controller based outside the EU.

  Penalties for GDPR non-compliance

  Mishandling sensitive data (or processing it when you’re not allowed to) can result in huge penalties. There are two tiers of GDPR fines :

  • €10 million or 2% of a company’s annual revenue for less severe infringements
  • €20 million or 4% of a company’s annual revenue for more severe infringements

  In the first half of 2023 alone, fines imposed in the EU due to GDPR violations exceeded €1.6 billion, up from €73 million in 2019.

  Examples of high-profile violations in the last few years include :

  • Amazon : The Luxembourg National Commission fined the retail giant with a massive $887 million fine in 2021 for not processing personal data per the GDPR. 
  • Google : The National Data Protection Commission (CNIL) fined Google €50 million for not getting proper consent to display personalised ads.
  • H&M : The Hamburg Commissioner for Data Protection and Freedom of Information hit the multinational clothing company with a €35.3 million fine in 2020 for unlawfully gathering and storing employees’ data in its service centre.

  One of the criteria that affects the severity of a fine is “data category” — the type of personal data being processed. Companies need to take extra precautions with sensitive data, or they risk receiving more severe penalties.

  What’s more, GDPR violations can negatively affect your brand’s reputation and cause you to lose business opportunities from consumers concerned about your data practices. 76% of consumers indicated they wouldn’t buy from companies they don’t trust with their personal data.

  Organisations should lay out their data practices in simple terms and make this information easily accessible so customers know how their data is being handled.

  Get started with GDPR-compliant web analytics

  The GDPR offers a framework for securing and protecting personal data. But it also distinguishes between sensitive and non-sensitive data. Understanding these differences and applying the lawful basis for processing this data type will help ensure compliance.

  Looking for a GDPR-compliant web analytics solution ?

  At Matomo, we take data privacy seriously. 

  Our platform ensures 100% data ownership, putting you in complete control of your data. Unlike other web analytics solutions, your data remains solely yours and isn’t sold or auctioned off to advertisers. 

  Additionally, with Matomo, you can be confident in the accuracy of the insights you receive, as we provide reliable, unsampled data.

  Matomo also fully complies with GDPR and other data privacy laws like CCPA, LGPD and more.

  Start your 21-day free trial today ; no credit card required. 

  Disclaimer

  We are not lawyers and don’t claim to be. The information provided here is to help give an introduction to GDPR. We encourage every business and website to take data privacy seriously and discuss these issues with your lawyer if you have any concerns.

  Try Matomo for Free

  21 day free trial. No credit card required.

  
  

  Your email address
  

  Your website address
  

  Your Analytics subdomain will be .matomo.cloud (change)

  Choose your analytics subdomain
  

  .matomo.cloud

  I accept the terms and conditions and data processing agreement (DPA)

  
  Your information will be used to create an account on our cloud service. For more information please consult our privacy policy.
  
  
  

  Start improving your websites now