Conducting Audio Forensics with Audacity
Introduction to Audio Forensics
Audio forensics is the scientific examination and analysis of audio recordings to determine their authenticity, enhance intelligibility, identify speakers, or extract evidence for legal proceedings. In the Philippines, where audio recordings are increasingly used as evidence in courts, law enforcement investigations, and corporate disputes, understanding audio forensics has become crucial for computer engineers, IT professionals, law enforcement personnel, and legal practitioners.
This comprehensive guide will teach you how to conduct audio forensics using Audacity, a free and open-source digital audio editor. Whether you’re analyzing recordings for legal evidence, investigating audio authenticity, or enhancing clarity for investigative purposes, this tutorial provides practical techniques grounded in Philippine legal standards and international forensic best practices.
Legal Framework: Audio Evidence in the Philippines
Rules on Electronic Evidence
In the Philippines, audio recordings fall under the category of electronic evidence governed by the Rules on Electronic Evidence (A.M. No. 01-7-01-SC) promulgated by the Supreme Court. The Revised Rules on Evidence, as amended in 2019 and effective May 1, 2020, explicitly recognizes electronic documents including “words, sounds, numbers, or their equivalent recorded in a retrievable form.”
Under Republic Act No. 8792 (the Electronic Commerce Act of 2000), audio recordings are admissible as evidence provided they meet the requirements for:
- Relevance – The audio must be relevant to the facts in issue
- Authentication – The proponent must prove that the audio is what it purports to be
- Integrity – The recording must be shown to be unaltered or, if altered, that alterations are documented
- Reliability – The manner of recording, storage, and handling must demonstrate trustworthiness
For computer engineers working with audio evidence, understanding these legal requirements is essential to ensure your forensic analysis meets admissibility standards in Philippine courts.
Chain of Custody
The chain of custody refers to the documented process showing who had access to evidence from the moment it was obtained to its presentation in court. For audio forensics, you must maintain:
- Date and time of acquisition
- Identity of person who collected the recording
- Description of recording device and settings
- Log of all persons who handled or analyzed the file
- Hash values (MD5 or SHA-256) of original files to prove integrity
- Documentation of all forensic procedures performed
Best Practice: Always work on copies of the original audio file. Never modify the original recording. Store the original in a secure location with restricted access.
Getting Started with Audacity
System Requirements and Installation
Minimum Requirements:
- Windows 10/11, macOS 10.13+, or Linux
- 2 GB RAM (4 GB recommended for large files)
- 64-bit operating system recommended
- Sufficient storage for working files (audio files can be large)
Installation Steps:
- Download Audacity from the official website: https://www.audacityteam.org/
- For Windows users in the Philippines, download the Windows 64-bit installer
- Run the installer and follow the setup wizard
- Install the optional LAME MP3 encoder and FFmpeg library for additional format support
Understanding the Audacity Interface
Before conducting forensic analysis, familiarize yourself with these key interface elements:
- Menu Bar – Contains File, Edit, View, Transport, Tracks, Generate, Effect, Analyze, Tools, and Help menus
- Transport Toolbar – Playback controls (Play, Pause, Stop, Skip to Start/End, Record)
- Tools Toolbar – Selection Tool, Envelope Tool, Draw Tool, Zoom Tool, Time Shift Tool, Multi-Tool
- Edit Toolbar – Cut, Copy, Paste, Trim, Silence
- Track Control Panel – Controls for each audio track (mute, solo, track dropdown menu)
- Waveform Display – Visual representation of audio amplitude over time
- Timeline – Shows time markers and selection boundaries
Essential Audio Forensics Workflow
Step 1: Secure the Original Recording
Before opening any audio file in Audacity:
- Create a hash value of the original file using command-line tools:
- Windows:
certutil -hashfile audiofile.wav SHA256 - Linux/Mac:
sha256sum audiofile.wav
- Windows:
- Document metadata including file size, creation date, modification date, and file format
- Make a working copy of the file and store the original in a secure, write-protected location
- Record these details in your forensic case file or evidence log
Step 2: Import and Initial Assessment
- Open Audacity and go to
File > Import > Audio - Select your working copy of the audio file
- Audacity will display the waveform in the track window
- Note the following technical parameters (shown in the track dropdown menu):
- Sample Rate (e.g., 44100 Hz, 48000 Hz) – Higher rates provide better quality
- Bit Depth (e.g., 16-bit, 24-bit, 32-bit float) – Higher depth means greater dynamic range
- Audio Format (Mono or Stereo)
- Duration of the recording
Forensic Note: Original recordings from smartphones typically use AAC format at 44.1 kHz sample rate. Voice recorders may use lower rates like 22.05 kHz or 16 kHz. Inconsistencies in these parameters can indicate editing or file conversion.
Step 3: Waveform Analysis
The waveform view shows audio amplitude (volume) over time. Look for:
- Abrupt discontinuities – Sharp, unnatural breaks in the waveform may indicate edits or splices
- Clipping – Flat-topped waveforms at maximum amplitude indicate distortion from recording too loud
- Silence periods – Unnatural silence or sections with zero amplitude may indicate deletions
- Background noise patterns – Sudden changes in background noise characteristics can reveal edits
- Amplitude anomalies – Sections with dramatically different volume levels may indicate spliced audio from different sources
Tip: Use View > Zoom In (Ctrl+1) to examine suspicious areas closely. Use View > Zoom Out (Ctrl+3) to see the overall structure.
Advanced Forensic Techniques
Spectral Analysis
Spectral analysis displays audio as a spectrogram showing frequency content over time. This is one of the most powerful tools for detecting audio manipulation.
To enable Spectrogram view:
- Click the track dropdown menu (small arrow next to track name)
- Select Spectrogram or Spectrogram (log f)
- Adjust Spectrogram Settings from the same menu for better visualization
What to look for in a spectrogram:
- Frequency cutoffs – Most consumer devices cut frequencies above 15-16 kHz. If portions of audio extend beyond this while others don’t, it may indicate splicing from different sources
- Noise floor consistency – The background “noise carpet” should be consistent throughout. Changes in noise floor patterns often reveal edits
- Unnatural patterns – Perfectly straight lines, grid patterns, or geometric shapes can indicate digital processing or synthetic audio
- ENF traces – Faint horizontal lines at 60 Hz (or 50 Hz in some regions) represent electrical network frequency, useful for authentication
- Compression artifacts – MP3 compression creates characteristic patterns. Multiple generations of compression create distinctive “banding” at high frequencies
Plot Spectrum Analysis
For detailed frequency distribution analysis:
- Select a portion of audio using the Selection Tool
- Go to
Analyze > Plot Spectrum - Choose analysis parameters:
- Algorithm: Spectrum (default for most forensic work)
- Function: Hann window (good balance)
- Size: 2048 or 4096 for detailed analysis
The Plot Spectrum window displays frequency (Hz) on the X-axis and amplitude (dB) on the Y-axis. Compare spectrums from different parts of the recording to identify inconsistencies.
ENF (Electrical Network Frequency) Analysis
ENF analysis is a powerful authentication technique that exploits the fact that electrical grid frequency (60 Hz in the Philippines, following North American standards) fluctuates slightly over time. These fluctuations are recorded in audio as a faint hum from electrical equipment.
Why ENF matters: Since grid frequency fluctuations are unique to specific times and locations, they serve as a “timestamp fingerprint” that can verify when and where a recording was made, or detect if audio has been spliced from different times.
Basic ENF analysis in Audacity:
- Switch to Spectrogram view
- Look for a horizontal line at approximately 60 Hz (Philippine grid frequency)
- Zoom in on the frequency axis to see variations in this line
- Discontinuities or jumps in the ENF trace may indicate edits or splicing
Note: Full ENF analysis requires specialized software and access to grid frequency databases. Audacity can only provide preliminary visual inspection. For court-admissible ENF analysis, consult a certified audio forensic expert.
Metadata Examination
Audio file metadata can reveal crucial information about recording devices, software, and editing history. While Audacity does not display all metadata, you can use external tools:
Windows:
- Right-click file → Properties → Details tab
- Use MediaInfo (free tool) for comprehensive metadata
Command-line tools:
- ExifTool – Comprehensive metadata extraction:
exiftool audiofile.m4a - FFprobe (part of FFmpeg):
ffprobe -show_format -show_streams audiofile.mp3
Key metadata fields to examine:
- Creation Date/Time – When the file was originally created
- Modification Date/Time – When the file was last modified (should match creation time if unedited)
- Software/Application – Recording or editing software used
- Device Information – Recording device model (e.g., iPhone 12, Samsung Galaxy S21)
- Encoder – Audio codec used
- GPS Coordinates (if available) – Location where recording was made
Forensic Red Flags:
- Modification date differs from creation date
- Software listed is not typical for claimed recording device
- Missing metadata that should be present for device type
- Multiple encoding generations evident in metadata
Audio Enhancement Techniques
Noise Reduction
Noise reduction improves intelligibility by removing unwanted background sounds. This is often necessary for poor-quality recordings used as evidence.
Proper noise reduction procedure:
- Identify a noise profile:
- Find a section containing only background noise (no speech or desired audio)
- Select this section (typically 0.5-2 seconds)
- Go to
Effect > Noise Reduction - Click “Get Noise Profile”
- Apply noise reduction:
- Select the entire audio track (Ctrl+A)
- Go to
Effect > Noise Reductionagain - Adjust settings:
- Noise Reduction (dB): Start with 12-18 dB (too much causes artifacts)
- Sensitivity: 6.00 is default (increase to remove more noise)
- Frequency Smoothing: 3-6 bands
- Click “Preview” to test
- If acceptable, click “OK” to apply
Forensic Caution: Aggressive noise reduction can introduce artifacts or distort speech. For legal evidence, use conservative settings and document all parameters used. Consider providing both original and enhanced versions.
Normalization and Amplification
Low-volume recordings may need amplification for better analysis:
- Normalize (
Effect > Normalize): Automatically adjusts amplitude to a target level (default -1.0 dB) without clipping. Use for overall volume boost. - Amplify (
Effect > Amplify): Manual volume adjustment. Shows maximum amplification possible without clipping. Use for precise control. - Compressor (
Effect > Compressor): Reduces dynamic range by making quiet parts louder and loud parts quieter. Useful for recordings with inconsistent volume.
Equalization (EQ)
EQ adjusts specific frequency ranges to improve clarity:
- Select audio section to enhance
- Go to
Effect > Filter Curve EQ(or legacyEqualization) - For speech enhancement:
- Boost 2-4 kHz range (speech intelligibility)
- Cut below 80 Hz (remove rumble)
- Cut above 8 kHz (reduce hiss) if needed
Forensic Documentation: Save EQ curve settings and include them in your forensic report. This allows verification and replication of your enhancement process.
Click and Pop Removal
Older recordings or poor connections may contain clicks and pops:
- Select the affected audio
- Go to
Effect > Click Removal - Adjust threshold and spike width
- Preview and apply if satisfactory
Detecting Audio Manipulation
Visual Indicators of Editing
In Waveform View:
- Sharp discontinuities in the waveform at edit points
- Sudden changes in background noise level or character
- Unnatural silence (complete zero amplitude) between segments
- Mismatched amplitude envelopes at splice points
In Spectrogram View:
- Vertical discontinuities in frequency content
- Changes in noise floor color/intensity
- Inconsistent high-frequency cutoff points
- Different compression artifacts in different sections
- Breaks in the ENF trace (if visible)
Speed and Pitch Manipulation Detection
Audio may be altered to change the speed or pitch of speech:
- Simple speed change: Both pitch and tempo change together. Sounds unnatural (chipmunk effect or slow motion)
- Time stretching: Tempo changes without affecting pitch. May introduce artifacts detectable in spectrograms
- Pitch shifting: Pitch changes without affecting tempo. Can be detected by analyzing formant frequencies
Detection method:
- Select a speech portion
- Use
Analyze > Plot Spectrum - Look for unnatural formant spacing or spectral anomalies
- Compare with known authentic speech samples from the same speaker
Copy-Paste Detection
Look for identical background noise patterns or repeated audio segments:
- Identify suspicious sections
- Select and copy suspected duplicate segment
- Create a new track and paste it
- Align with original using Time Shift Tool
- Invert one track (
Effect > Invert) - If tracks are identical, they will cancel out completely (null test)
Forensic Documentation and Reporting
Essential Documentation
A complete forensic audio analysis report should include:
- Case Information
- Case number and date
- Client or requesting party
- Examiner name and qualifications
- Evidence Description
- File name, format, and size
- Date received and source
- Chain of custody documentation
- Hash values (MD5/SHA-256) of original files
- Technical Parameters
- Sample rate, bit depth, channels
- Duration and file format
- Recording device (if known)
- Metadata summary
- Analysis Methodology
- Software used (Audacity version)
- Analytical techniques applied
- Settings and parameters for all processes
- Findings
- Observations from waveform analysis
- Spectral analysis results
- Evidence of editing or manipulation
- Assessment of authenticity
- Identified speakers (if applicable)
- Visual Evidence
- Screenshots of relevant waveforms
- Spectrograms showing anomalies
- Annotated images highlighting findings
- Conclusions
- Summary of findings
- Opinion on authenticity (if qualified to provide)
- Limitations of analysis
- Attachments
- Processed audio files (with documentation of all modifications)
- Log files from analysis software
- Enhanced versions created
Exporting Results from Audacity
For evidence preservation:
- Save project:
File > Save Project > Save Project(creates .aup3 file with all edits and tracks) - Export enhanced audio:
File > Export > Export Audio- Choose WAV format (uncompressed, lossless) for forensic use
- Use maximum quality settings
- Document export settings
- Capture screenshots: Use Audacity’s built-in screenshot function or operating system tools to capture relevant visual evidence
Practical Exercise: Analyzing a Sample Recording
Exercise Setup
To practice these forensic techniques, create or obtain a test audio recording. For educational purposes, you can:
- Record a voice memo on your smartphone
- Add background noise (fan, traffic, music)
- Create a second recording with different background
- Use Audacity to splice portions together
- Export as MP3
- Now analyze this “evidence” as if it were submitted for forensic examination
Analysis Checklist
- ☐ Document original file hash and metadata
- ☐ Import into Audacity and note technical parameters
- ☐ Perform initial waveform inspection
- ☐ Identify any obvious discontinuities
- ☐ Switch to spectrogram view
- ☐ Look for frequency and noise floor inconsistencies
- ☐ Use Plot Spectrum on suspicious sections
- ☐ Check for ENF traces
- ☐ Apply noise reduction if needed
- ☐ Document all findings with screenshots
- ☐ Write a brief analysis report
Common Issues and Troubleshooting
Problem: “FFmpeg library not found”
Solution: Some audio formats (M4A, AAC, WMA) require the FFmpeg library.
- Download FFmpeg from the Audacity website
- Install or extract to a known location
- In Audacity, go to
Edit > Preferences > Libraries - Click “Locate” next to FFmpeg Library and point to the installation
Problem: Cannot import audio file
Possible causes:
- File is corrupted
- Unsupported format (requires FFmpeg or LAME)
- File is DRM-protected
- Insufficient disk space or memory
Solutions:
- Try converting file to WAV format using another tool first
- Check file integrity (try opening in another audio player)
- Free up disk space
- Install required codec libraries
Problem: Audio sounds distorted after processing
Causes:
- Over-amplification causing clipping
- Excessive noise reduction settings
- Multiple generations of lossy compression
Solutions:
- Use
Edit > Undo(Ctrl+Z) to revert changes - Apply effects with more conservative settings
- Always preview effects before applying to entire track
- Keep backups of processing stages
Problem: Spectrogram shows no detail
Solution: Adjust spectrogram settings:
- Click track dropdown menu
- Select “Spectrogram Settings”
- Try different settings:
- Increase window size (2048, 4096, 8192)
- Change scale to “Logarithmic”
- Adjust color scheme (“Grayscale” often shows more detail)
- Modify frequency range to focus on relevant frequencies
Best Practices for Audio Forensics
- Never modify the original: Always work on copies. Store originals in write-protected media.
- Document everything: Keep detailed notes of all procedures, settings, and observations.
- Use hash values: Calculate and record SHA-256 hashes before and after each processing step.
- Maintain chain of custody: Log who accessed files, when, and what was done.
- Be conservative with enhancements: Over-processing can create artifacts and reduce evidential value.
- Provide both versions: Supply original and enhanced versions with full documentation of changes.
- Know your limits: Recognize when analysis requires specialized tools or expert testimony beyond basic Audacity capabilities.
- Stay objective: Report findings accurately without bias toward desired outcomes.
- Follow legal requirements: Ensure your methods comply with Philippine Rules on Electronic Evidence.
- Keep learning: Audio forensics is a rapidly evolving field. Stay updated on new techniques and tools.
Real-World Applications in the Philippines
Law Enforcement Investigations
Philippine law enforcement agencies use audio forensics for:
- Enhancing recorded confessions for clarity
- Analyzing wiretap recordings in criminal cases
- Verifying authenticity of evidence recordings
- Speaker identification in threat calls or ransom demands
Civil and Criminal Litigation
Audio forensics supports legal cases involving:
- Contract disputes (analyzing recorded agreements)
- Employment cases (wrongful termination, harassment claims)
- Intellectual property theft (voice identification)
- Defamation cases involving audio recordings
Corporate Investigations
- Employee misconduct investigations
- Fraud detection through call recordings
- Verification of business communications
- Compliance monitoring for financial institutions
Journalism and Media Verification
- Verifying authenticity of leaked audio
- Detecting deepfake audio
- Enhancing interview recordings
- Analyzing public official statements for authenticity
Frequently Asked Questions (FAQ)
Is Audacity sufficient for legal evidence analysis in Philippine courts?
Audacity is a legitimate tool for preliminary audio forensic analysis and can be used for basic enhancement and examination. However, for court-admissible expert testimony, Philippine courts generally prefer analysis conducted by certified forensic audio experts using validated forensic software suites. Audacity results can support other evidence but may not carry the same weight as professional forensic analysis, especially in contested cases.
Can I use Audacity for voice identification?
Audacity can assist in preliminary voice comparison through visual waveform and spectral analysis, but definitive voice identification requires specialized voice biometric software and expert analysis. Philippine courts require qualified expert testimony for voice identification evidence.
How do I prove that audio hasn’t been edited?
Proving authenticity requires multiple layers of evidence: (1) hash values showing file integrity from acquisition to analysis, (2) metadata examination showing no modification, (3) waveform and spectral analysis showing no discontinuities, (4) ENF analysis matching grid frequency records for claimed time/location, (5) chain of custody documentation, and (6) examination by qualified expert if challenged in court.
What format should I use for exporting enhanced audio?
For forensic purposes, always export in WAV format (uncompressed, 16-bit or 24-bit, same sample rate as original). This preserves maximum audio quality without introducing compression artifacts. Provide both the original unmodified file and the enhanced version, clearly labeled, with documentation of all processing steps.
Is it legal to record conversations in the Philippines?
The legality of recording conversations in the Philippines depends on the context:
- One-party consent: Generally legal if you are a participant in the conversation
- Third-party recordings: Recording others’ private conversations without consent may violate RA 4200 (Anti-Wiretapping Act)
- Admissibility: Even if legally obtained, recordings must meet authentication requirements under Rules on Electronic Evidence
Consult legal counsel for specific situations.
Can I detect AI-generated or deepfake audio with Audacity?
Audacity can help identify some indicators of synthetic audio (unnatural spectral patterns, perfect consistency in background noise, artifacts at splice points), but sophisticated deepfakes require specialized AI detection tools and expert analysis. Look for: perfectly clean noise floor, unnatural formant frequencies, missing breathing sounds, and spectral anomalies in high frequencies.
What qualifications do I need to be an audio forensic expert in the Philippines?
While there’s no single certification requirement, forensic audio experts typically have:
- Degree in acoustics, audio engineering, computer engineering, or related field
- Specialized training in audio forensics
- Experience with professional forensic software
- Understanding of Philippine evidence rules
- International certifications (e.g., from Audio Engineering Society or International Association for Identification) strengthen credibility
How long should I preserve forensic case files?
For legal cases, preserve all files (originals, working copies, enhanced versions, project files, reports, and documentation) until the statute of limitations expires or appeals are exhausted. Best practice is to maintain backups for at least 10 years for civil cases, longer for criminal cases. Store on multiple media in secure locations.
Can background music copyright be an issue in forensic analysis?
For forensic analysis purposes, using copyrighted material as evidence falls under fair use doctrine. However, if you’re publishing or distributing enhanced recordings beyond court proceedings, copyright issues may arise. For legal proceedings, focus on documenting the analysis rather than distributing the audio itself.
What should I do if I find evidence of tampering?
If your analysis reveals evidence of audio manipulation:
- Document all indicators of tampering with screenshots and detailed notes
- Preserve your analysis project file
- Write a detailed report of findings
- Recommend engaging a certified forensic audio expert for confirmation
- Notify relevant parties (legal counsel, investigator, court) as appropriate
Additional Resources
Software and Tools
- Audacity Manual: https://manual.audacityteam.org/
- Audacity Forum: Community support for technical questions
- MediaInfo: Free tool for detailed audio metadata inspection
- ExifTool: Comprehensive metadata extraction tool
Philippine Legal Resources
- Rules on Electronic Evidence: Available at Chan Robles Virtual Law Library
- RA 8792 (E-Commerce Act): Official Gazette
- RA 4200 (Anti-Wiretapping Act): LawPhil.net
- Supreme Court E-Library: elibrary.judiciary.gov.ph
Professional Development
- Audio Engineering Society (AES): Offers resources and standards for audio forensics
- European Network of Forensic Science Institutes (ENFSI): Publishes best practice guidelines for ENF analysis and forensic audio
- International Association for Identification (IAI): Certifications in forensic audio analysis
Recommended Reading
- “Forensic Audio Analysis: Techniques and Tools for the Modern Investigator” – Eclipse Forensics
- “Audio Forensics: Theory and Practice” – Academic resources on spectral and ENF analysis
- “Proper Authentication of Electronic Evidence in Philippine Courts” – Respicio & Co. legal analysis
Conclusion
Audio forensics with Audacity provides computer engineers, IT professionals, and investigators in the Philippines with powerful capabilities for analyzing and enhancing audio evidence. While Audacity is an excellent entry-level tool for preliminary analysis, understanding its limitations is crucial—complex cases requiring court testimony typically need professional forensic audio analysis using validated commercial software.
The key to effective audio forensics is mastering both technical skills and legal requirements under Philippine law. Always prioritize:
- Evidence integrity through proper chain of custody
- Thorough documentation of all procedures
- Objective analysis free from bias
- Compliance with Rules on Electronic Evidence
- Continuous learning as forensic techniques evolve
Whether you’re enhancing a crucial witness recording, verifying the authenticity of audio evidence, or conducting preliminary analysis for law enforcement, the techniques in this guide provide a solid foundation. Practice regularly, stay updated on emerging deepfake technologies and AI-generated audio, and know when to engage certified experts for cases requiring court-admissible testimony.
Remember: the ultimate goal of audio forensics is not just technical mastery, but serving justice through accurate, reliable analysis that stands up to legal scrutiny.
References
- Rules on Electronic Evidence (A.M. No. 01-7-01-SC), Supreme Court of the Philippines. Retrieved from https://www.chanrobles.com/rulesonelectronicevidence.htm
- Republic Act No. 8792 (Electronic Commerce Act of 2000). Retrieved from Supreme Court E-Library: https://elibrary.judiciary.gov.ph/
- 2019 Amendments to the 1989 Rules on Evidence (A.M. No. 19-08-15-SC). Retrieved from LawPhil.net: https://lawphil.net/courts/rules/am_19-08-15-sc_2019.html
- “Proper Authentication of Electronic Evidence in Philippine Courts,” Respicio & Co. Retrieved from https://www.respicio.ph/commentaries/proper-authentication-of-electronic-evidence-in-philippine-courts
- “Forensic Audio Authentication: Techniques and Tools for the Modern Investigator,” Eclipse Forensics (2024). Retrieved from https://eclipseforensics.com/
- “Best Practice Guidelines for ENF Analysis in Forensic Authentication of Digital Evidence,” European Network of Forensic Science Institutes (ENFSI) Forensic Speech and Audio Analysis Working Group. Retrieved from https://enfsi.eu/
- “Spectral Analysis,” Audacity Support Documentation. Retrieved from https://support.audacityteam.org/audio-analysis/spectral-analysis
- Audacity Manual, Audacity Team. Retrieved from https://manual.audacityteam.org/
- “Audio Forensics Methods & Standards: Practical Guide,” Audio Forensic Expert (2024). Retrieved from https://www.audioforensicexpert.eu/
Disclaimer: This tutorial is for educational purposes. Audio forensic analysis for legal proceedings should be conducted by qualified experts with appropriate certifications. The author and publisher assume no liability for the use or misuse of information contained herein. When in doubt, consult legal counsel and certified forensic audio experts.
Last Updated: November 2025
