SRTGen.com

Ai Transcription

Explore Ai Transcription on SRTGen.com — AI Subtitles, Dubbing, and Voice Studio.

Convert Audio & Video to Accurate Text

Generate fast, accurate AI transcriptions for videos, podcasts, meetings, interviews, lectures, and recordings. Export TXT, SRT, VTT, ASS, and more in one click.

Speech to Text in One Click

Convert any video or audio file into structured transcripts in three simple steps.

Built for Speed and Precision

Discover the powerful suite of transcription capabilities engineered to streamline your audio and video translation workflows.

State-of-the-Art Accuracy

Powered by world-class speech AI models that deliver flawless transcription across 100+ languages, handling even the most complex accents, low-quality recordings, and noisy environments with absolute precision.

Full RTL Support

Flawless rendering and alignment for Right-to-Left languages including Arabic, Hebrew, and Persian in subtitle edits and text files.

Translate & Easy Edit

Translate your transcription into 100+ languages in seconds, and make quick refinements on the fly with our side-by-side comparison editor.

Certified Human QA

Need absolute precision? Delegate transcripts to our native expert linguists to guarantee 99%+ accuracy SLA under SOC-2 confidentiality.

Frequently Asked Questions

Everything you need to know about SRTGen's AI transcription.

Work with the transcription experts

Join over 50,000+ businesses, researchers, and professional creators who trust SRTGen for high-precision, secure transcripts.

Convert Audio & Video to Accurate Text

Generate fast, accurate AI transcriptions for videos, podcasts, meetings, interviews, lectures, and recordings. Export TXT, SRT, VTT, ASS, and more in one click.

Speech to Text in One Click

Convert any video or audio file into structured transcripts in three simple steps.

Built for Speed and Precision

Discover the powerful suite of transcription capabilities engineered to streamline your audio and video translation workflows.

State-of-the-Art Accuracy

Powered by world-class speech AI models that deliver flawless transcription across 100+ languages, handling even the most complex accents, low-quality recordings, and noisy environments with absolute precision.

Full RTL Support

Flawless rendering and alignment for Right-to-Left languages including Arabic, Hebrew, and Persian in subtitle edits and text files.

Translate & Easy Edit

Translate your transcription into 100+ languages in seconds, and make quick refinements on the fly with our side-by-side comparison editor.

Certified Human QA

Need absolute precision? Delegate transcripts to our native expert linguists to guarantee 99%+ accuracy SLA under SOC-2 confidentiality.

Frequently Asked Questions

Everything you need to know about SRTGen's AI transcription.

Work with the transcription experts

Join over 50,000+ businesses, researchers, and professional creators who trust SRTGen for high-precision, secure transcripts.

Every Scale Needs Powerful Tools

Powerful AI-driven capabilities designed to make your subtitling workflow 10x faster.

Word-level Precision Editing

Our deep learning models capture every word with up to 99% accuracy, even with background noise or music.

Break Barriers with RTL Support

Instantly translate your subtitles into 50+ languages with full support for RTL and special characters.

High Quality & Low Prices

Get professional-grade results for a fraction of the cost. No hidden fees, just honest, scalable pricing.

Tag @SRTGenDotCom on X

Connect your social accounts and let our AI automatically generate and post subtitles for your Reels, Shorts, and TikToks.

Everything you need for Subtitles

Powerful tools designed for content creators, agencies, and global brands.

Viral Video Captions

Style your videos with dynamic, animated captions similar to top influencers like Alex Hormozi.

Global Subtitle Translation

Break language barriers by instantly translating your content into 50+ languages with high accuracy.

AI Note Taker & Summary

Actionable insights in seconds.

Team Collaboration

Work together with your team in real-time.

Social Automation

Schedule and post your captioned videos directly to TikTok, Instagram, and YouTube with a single click.

Pro Subtitle Editor

A powerful timeline-based editor to perfectly sync every word, adjust styles, and add custom animations.

Audio to Video

Convert podcasts or voice recordings into engaging social animations with waveforms and captions.

Watch Without Sound

Up to 85% of social video views happen with the sound off. Furthermore, 69% of consumers watch muted in public places.

Global Viewers

On average, over two-thirds of a creator's audience watch time comes from outside of their home area.

Americans Prefer Subtitles

Over 50% of Americans prefer watching content with subtitles, a trend led by 70% of Gen Z viewers.

Non-English Preferences

76% of online consumers prefer to buy products in their native language, and 40% will never buy from websites in other languages.

Subtitles Boost Retention

Viewers are 80% more likely to watch a video to the very end if subtitles or captions are available.

Native Language Preference

80% of international viewers prefer watching video content in their own native language, making localization critical for global reach.

Watch Time Increase

Videos with closed captions and subtitles experience an average increase in watch time of 12% to 25%.

Real-Time Processing

AI transcription processes audio at 3-5x real-time speed, converting hours of meetings or interviews into text in minutes.

Manual Effort Required

Human transcriptionists typically require 4 to 6 hours of work to transcribe just 1 single hour of audio or video.

Transcription Cost Cut

AI transcription reduces costs to as low as $0.01 per minute, compared to $1.50+ per minute for manual transcription services.

Localization Savings

Traditional dubbing agencies charge $50–$150+ per minute, whereas AI voice cloning dubbing starts at under $0.50 per minute.

Podcast Drop-off Rate

Approximately 19% of audio content listeners will instantly stop listening to a show due to poor audio quality or noise.

Low Audio, Low Trust

Scientific research shows that poor audio quality significantly reduces how intelligent, likeable, and trustworthy a speaker is perceived to be.

Cognitive Fatigue Drop

Excess background noise forces the listener's brain to work harder, causing most listeners to drop off within 2 minutes.

Global South Growth

More than 60% of the world's Christians live in the Global South (Africa, Asia, and Latin America). Translating sermons and prayers is essential for global outreach.

Global Christian Faith

Over 2.6 billion people identify as Christian globally. Translating prayers, scripture audio, and worship videos bridges linguistic gaps across diverse faith communities.

Hearing Loss Impact

Over 1.5 billion people (20% of the global population) live with hearing loss. Without captions on videos, they miss key parts of the message.

International e-Learners

Over 80% of top online learning platforms' students reside outside the United States. Localizing courses expands your addressable student base by up to 5x.

Higher Comprehension

Studies on multimedia learning show that adding subtitles and voiceovers to educational videos boosts material comprehension by over 30% for diverse students.

Preference for Video

83% of employees prefer to consume training and informational content via video rather than text-only documentation.

Online News Video Reach

77% of online consumers globally watch digital news videos each week. Translating and subtitling reports allows newsrooms to tap into global markets.

Social News Consumption

54% of global audiences access news via social networks and video platforms. Adding timed captions is key to capturing silent social feeds.

Muted Mobile Views

92% of online consumers watch videos with the sound off on mobile devices. Captions are crucial to deliver news stories without audio.

Weekly Time Savings

Over 62% of corporate professionals save more than 4 hours weekly by using automated AI transcription for meeting logs and action items.

Communication Gaps

67% of global employees collaborate with team members who speak different primary languages, highlighting the need for translated communications.

Productivity Increase

Standardized video onboarding and localized training materials increase new hire productivity by over 50%.

Inclusion & Accessibility

Over 80% of virtual event organizers report that providing real-time captions and transcripts significantly improves attendee retention and accessibility.

Unique Search Traffic Boost

Case studies on publishing transcripts of audio and video content show an average 6.68% increase in unique search engine visitors.

Higher Watch Times

Post-event on-demand replays localized with automated subtitles and voice dubbing experience a 40% boost in viewer watch time and completion rate.

AI Transcription Accuracy

Advanced AI models transcribe qualitative research interviews at over 95% accuracy under clean acoustic conditions.

Research Efficiency Gain

Qualitative researchers report saving up to 80% of transcription and coding time by using searchable AI transcripts instead of manual methods.

Global Focus Groups

Translating and captioning participant videos unlocks qualitative focus groups across 100+ language demographics, allowing seamless international insights.

Audiobook Listener Growth

Localizing audiobook catalogs into major global languages like Spanish, French, and German expands your listenership footprint by up to 3.5x.

Voice Production Savings

Using AI narration and voice cloning reduces the cost of multilingual voiceover production by over 95% compared to booking studio talent.

Natural Voice Retention

Synthesized audiobooks that keep natural breathing, inflection patterns, and emotional pacing see 45% higher retention than basic mechanical text-to-speech.

International Gaming Market

Over 70% of global gaming revenue is generated outside of North America. Localizing game trailers, store pages, and audio devlogs is key to global sales.

Character Dubbing Savings

AI voice cloning and automated dialogue dubbing reduce character voice production budgets by over 90% compared to traditional recording sessions.

Players Enable Subtitles

Over 80% of modern video game players keep subtitles enabled during gameplay and cinematic cutscenes for narrative clarity.

Viewership Growth

Adding localized subtitles and multilingual captions to product videos increases global viewership by up to 40%.

Consumer Purchase Intent

76% of online shoppers prefer to buy products with information available in their native language, and 40% will never buy from websites in other languages.

Higher Video Completion

Adding timed captions on marketing videos on LinkedIn, Facebook, and Instagram boosts video completion rates by 80%, ensuring your pitch is heard.

Transcription Boosts Completion

Providing transcripts alongside podcast audio increases overall episode completion rates by up to 80% for listeners.

Frequently Asked Questions

Find answers to the most common questions about SRTGen's AI, billing, and automation features.

Still need help?

Can't find the answer you're looking for? Reach out to our support team and we will get back to you soon.

Every Scale Needs Powerful Tools

Powerful AI-driven capabilities designed to make your subtitling workflow 10x faster.

Word-level Precision Editing

Our deep learning models capture every word with up to 99% accuracy, even with background noise or music.

Break Barriers with RTL Support

Instantly translate your subtitles into 50+ languages with full support for RTL and special characters.

High Quality & Low Prices

Get professional-grade results for a fraction of the cost. No hidden fees, just honest, scalable pricing.

Tag @SRTGenDotCom on X

Connect your social accounts and let our AI automatically generate and post subtitles for your Reels, Shorts, and TikToks.

Watch Without Sound

Up to 85% of social video views happen with the sound off. Furthermore, 69% of consumers watch muted in public places.

Global Viewers

On average, over two-thirds of a creator's audience watch time comes from outside of their home area.

Americans Prefer Subtitles

Over 50% of Americans prefer watching content with subtitles, a trend led by 70% of Gen Z viewers.

Non-English Preferences

76% of online consumers prefer to buy products in their native language, and 40% will never buy from websites in other languages.

Subtitles Boost Retention

Viewers are 80% more likely to watch a video to the very end if subtitles or captions are available.

Native Language Preference

80% of international viewers prefer watching video content in their own native language, making localization critical for global reach.

Watch Time Increase

Videos with closed captions and subtitles experience an average increase in watch time of 12% to 25%.

Real-Time Processing

AI transcription processes audio at 3-5x real-time speed, converting hours of meetings or interviews into text in minutes.

Manual Effort Required

Human transcriptionists typically require 4 to 6 hours of work to transcribe just 1 single hour of audio or video.

Transcription Cost Cut

AI transcription reduces costs to as low as $0.01 per minute, compared to $1.50+ per minute for manual transcription services.

Localization Savings

Traditional dubbing agencies charge $50–$150+ per minute, whereas AI voice cloning dubbing starts at under $0.50 per minute.

Podcast Drop-off Rate

Approximately 19% of audio content listeners will instantly stop listening to a show due to poor audio quality or noise.

Low Audio, Low Trust

Scientific research shows that poor audio quality significantly reduces how intelligent, likeable, and trustworthy a speaker is perceived to be.

Cognitive Fatigue Drop

Excess background noise forces the listener's brain to work harder, causing most listeners to drop off within 2 minutes.

Global South Growth

More than 60% of the world's Christians live in the Global South (Africa, Asia, and Latin America). Translating sermons and prayers is essential for global outreach.

Global Christian Faith

Over 2.6 billion people identify as Christian globally. Translating prayers, scripture audio, and worship videos bridges linguistic gaps across diverse faith communities.

Hearing Loss Impact

Over 1.5 billion people (20% of the global population) live with hearing loss. Without captions on videos, they miss key parts of the message.

International e-Learners

Over 80% of top online learning platforms' students reside outside the United States. Localizing courses expands your addressable student base by up to 5x.

Higher Comprehension

Studies on multimedia learning show that adding subtitles and voiceovers to educational videos boosts material comprehension by over 30% for diverse students.

Preference for Video

83% of employees prefer to consume training and informational content via video rather than text-only documentation.

Online News Video Reach

77% of online consumers globally watch digital news videos each week. Translating and subtitling reports allows newsrooms to tap into global markets.

Social News Consumption

54% of global audiences access news via social networks and video platforms. Adding timed captions is key to capturing silent social feeds.

Muted Mobile Views

92% of online consumers watch videos with the sound off on mobile devices. Captions are crucial to deliver news stories without audio.

Weekly Time Savings

Over 62% of corporate professionals save more than 4 hours weekly by using automated AI transcription for meeting logs and action items.

Communication Gaps

67% of global employees collaborate with team members who speak different primary languages, highlighting the need for translated communications.

Productivity Increase

Standardized video onboarding and localized training materials increase new hire productivity by over 50%.

Inclusion & Accessibility

Over 80% of virtual event organizers report that providing real-time captions and transcripts significantly improves attendee retention and accessibility.

Unique Search Traffic Boost

Case studies on publishing transcripts of audio and video content show an average 6.68% increase in unique search engine visitors.

Higher Watch Times

Post-event on-demand replays localized with automated subtitles and voice dubbing experience a 40% boost in viewer watch time and completion rate.

AI Transcription Accuracy

Advanced AI models transcribe qualitative research interviews at over 95% accuracy under clean acoustic conditions.

Research Efficiency Gain

Qualitative researchers report saving up to 80% of transcription and coding time by using searchable AI transcripts instead of manual methods.

Global Focus Groups

Translating and captioning participant videos unlocks qualitative focus groups across 100+ language demographics, allowing seamless international insights.

Audiobook Listener Growth

Localizing audiobook catalogs into major global languages like Spanish, French, and German expands your listenership footprint by up to 3.5x.

Voice Production Savings

Using AI narration and voice cloning reduces the cost of multilingual voiceover production by over 95% compared to booking studio talent.

Natural Voice Retention

Synthesized audiobooks that keep natural breathing, inflection patterns, and emotional pacing see 45% higher retention than basic mechanical text-to-speech.

International Gaming Market

Over 70% of global gaming revenue is generated outside of North America. Localizing game trailers, store pages, and audio devlogs is key to global sales.

Character Dubbing Savings

AI voice cloning and automated dialogue dubbing reduce character voice production budgets by over 90% compared to traditional recording sessions.

Players Enable Subtitles

Over 80% of modern video game players keep subtitles enabled during gameplay and cinematic cutscenes for narrative clarity.

Viewership Growth

Adding localized subtitles and multilingual captions to product videos increases global viewership by up to 40%.

Consumer Purchase Intent

76% of online shoppers prefer to buy products with information available in their native language, and 40% will never buy from websites in other languages.

Higher Video Completion

Adding timed captions on marketing videos on LinkedIn, Facebook, and Instagram boosts video completion rates by 80%, ensuring your pitch is heard.

Transcription Boosts Completion

Providing transcripts alongside podcast audio increases overall episode completion rates by up to 80% for listeners.

Trusted by 10,000+ Creators

> "The one-click dubbing is genuinely impressive. Upload, pick a language, done. The background music stays intact." > — **@Rivera** > "works great" > — **@Chen** > "We localize training videos into Spanish, French, and Japanese for global teams. One-click dubbing handles all three without any manual steps. The vocal separator keeps the original background audio clean." > — **@Thorne** > "good tool" > — **@Rossi** > "The emotion preservation is better than I expected. The dubbed voice doesn't sound flat or robotic — it actually keeps the pacing and pauses from the original." > — **@Wilson** > "I used my own cloned voice to dub a product video into French. Took about ten minutes total from clone to final export." > — **@Tanaka** > "The background music staying intact after dubbing is the thing that surprised me most. Other tools I tried would just drop it." > — **@Smith** > "really solid" > — **@Gomez** > "I do freelance video localization. Being able to use a cloned voice for dubbing instead of hiring voice talent for every language saves me a lot of budget." > — **@Harris** > "just works" > — **@Mason** > "The Speech Dubbing mode is great when I need more control over individual lines. One-click for bulk content, speech mode for detailed work." > — **@Lopez** > "100+ languages and the output actually sounds natural. Recommended." > — **@Patel** > "pretty good overall" > — **@Elena** > "I was worried the dubbed voice would sound off, but the tone and pacing match the original well. Clients haven't complained." > — **@Jack** > "Using a voice clone to dub into multiple languages is a game changer for solo creators. I don't need to hire different voice actors anymore." > — **@Amy** > "very useful" > — **@Brian** > "The vocal separator is what makes the output feel professional. Background music and effects stay exactly where they should be." > — **@Clara** > "Solid platform. I've been using it for about three months for client video work and it's been reliable." > — **@Dennis** > "does what it says" > — **@Evgeny** > "We use it to dub all our course videos. Spanish and Portuguese versions are up now. The quality is good enough that students haven't noticed it's AI-dubbed." > — **@Marie** > "The one-click flow is fast. Upload, choose language, export. That's it." > — **@Radu** > "very good" > — **@Franco** > "I run a small production studio. Having dubbing, voice cloning, and vocal separation all in one platform without managing multiple subscriptions makes a real difference to our workflow." > — **@Etta** > "solid and reliable" > — **@Lukas**

Frequently Asked Questions

How accurate is SRTGen's AI transcription?

Our SRTGen Pro engine achieves up to 96.9% accuracy (3.1% Word Error Rate) on clean English audio. It is powered by state-of-the-art speech models that excel at handling complex accents, fast speech patterns, and moderate background noise.

Does the AI separate and identify different speakers?

Yes, our transcription includes automatic speaker diarization. The AI recognizes when the speaker changes and labels them (e.g., Speaker 1, Speaker 2) with exact millisecond timestamps so you can easily associate text with the right voice.

Which file formats are supported for upload and export?

You can upload all common audio and video files, including MP4, MP3, WAV, M4A, MOV, and AAC. Once transcribed, you can export your transcript in formats like SRT, VTT, TXT, JSON, or CSV, ready for immediate use in video editors or document archives.

How long does it take to transcribe a file?

Transcribing is extremely fast. On average, a 1-hour audio file takes less than 2 to 3 minutes to transcribe completely, giving you near real-time results.

Can I translate my transcripts into other languages?

Absolutely. With our Translate & Easy Edit feature, you can translate your transcripts into over 100 languages instantly. The translated text remains perfectly synchronized with the source timestamps, allowing you to edit original and translated text side-by-side.

Is my uploaded audio/video data secure and private?

Yes, privacy is our top priority. We use AES-256 encryption at rest and SSL/TLS in transit, following SOC-2 security protocols. We strictly do not train our models on your files or data, ensuring your content remains completely confidential.

How accurate is SRTGen's AI transcription?

Our SRTGen Pro engine achieves up to 96.9% accuracy (3.1% Word Error Rate) on clean English audio. It is powered by state-of-the-art speech models that excel at handling complex accents, fast speech patterns, and moderate background noise.

Does the AI separate and identify different speakers?

Yes, our transcription includes automatic speaker diarization. The AI recognizes when the speaker changes and labels them (e.g., Speaker 1, Speaker 2) with exact millisecond timestamps so you can easily associate text with the right voice.

Which file formats are supported for upload and export?

You can upload all common audio and video files, including MP4, MP3, WAV, M4A, MOV, and AAC. Once transcribed, you can export your transcript in formats like SRT, VTT, TXT, JSON, or CSV, ready for immediate use in video editors or document archives.

How long does it take to transcribe a file?

Transcribing is extremely fast. On average, a 1-hour audio file takes less than 2 to 3 minutes to transcribe completely, giving you near real-time results.

Can I translate my transcripts into other languages?

Absolutely. With our Translate & Easy Edit feature, you can translate your transcripts into over 100 languages instantly. The translated text remains perfectly synchronized with the source timestamps, allowing you to edit original and translated text side-by-side.

Is my uploaded audio/video data secure and private?

Yes, privacy is our top priority. We use AES-256 encryption at rest and SSL/TLS in transit, following SOC-2 security protocols. We strictly do not train our models on your files or data, ensuring your content remains completely confidential.

What is SRTGen.com?

SRTGen is an AI-powered platform that automatically transcribes, translates, and styles subtitles for your short-form and long-form videos. It is designed to save you hours of manual editing.

Do I need to download any software?

No! SRTGen is completely cloud-based. You can upload videos directly from your browser or even paste social media links to import videos automatically.

Is there a free trial?

Yes, we offer a free tier that gives you 20 signup credits to test out our AI transcription, translation, and subtitle burning features before committing to a paid plan.

How accurate is the transcription?

Using state-of-the-art AI models, SRTGen delivers speech-to-text accuracy of up to 99% for clear audio. Our engines are highly resilient to background noise, music, and overlapping speakers.

How accurate is the AI translation?

We offer two levels of AI translation: Basic and Pro. For a detailed comparison of their features, accuracy levels, and pricing, please visit our Pricing Page.

Which languages are supported?

We support transcription in over 100 languages and instant translation in over 50 global languages. SRTGen features automatic language detection and can accurately process multiple languages within a single video segment.

Can I edit the generated subtitles?

Absolutely. Professional subtitle editing is what we specialize in. Our world-class editor allows you to correct words, adjust timings with frame-accurate precision, and split or merge segments with total ease.

Can I customize how the subtitles look?

Yes! You can choose from various viral and professional templates, or create your own by customizing fonts, colors, stroke, background, shadows, and word-by-word highlight animations.

What formats can I export?

You can download professional subtitle files (SRT, VTT, ASS, TXT) for use in Premiere or Final Cut. Additionally, you can burn subtitles into your videos directly in your browser locally for instant results, or use our Burn Video feature for high-performance cloud rendering.

Can I save my own custom templates?

Yes! Once you've perfected a subtitle style, you can save it as a preset. This allows you to apply your unique brand identity to any new video instantly, ensuring a consistent look across your social media channels.

What resolution and quality is the final video?

Our Burn Video engine supports high-fidelity rendering up to 4K resolution. We prioritize high bitrates and color accuracy, so your videos look pristine when uploaded to TikTok, Reels, or YouTube.

Is there a watermark on the videos?

No. We believe your content should be yours to keep. SRTGen does not add any watermarks to your burned videos, regardless of which plan you are on.

What is the @SRTGenDotCom on X?

It's our automated social bot. Once you connect your X account to your SRTGen profile, you can reply to any video on X tagging @SRTGenDotCom, and it will automatically reply with a newly subtitled version of that video.

How does the bot know what to do?

Our AI understands natural language. You can tweet '@SRTGenDotCom translate this to Spanish' or '@SRTGenDotCom add big yellow captions' and it figures out the rest automatically.

Does this cost extra?

No! Every plan comes with an included X (Twitter) quota that resets automatically every month. The number of allowed interactions per month depends on your chosen subscription level.

How long does it take for SRTGenDotCom to reply?

Typically less than 10 minutes. Our automated system polls for mentions every few minutes and processes them sequentially to ensure the best possible subtitle quality.

What if the tweet has multiple videos?

We process one video per request. To target a specific video, include the number in your tweet (e.g. "video 2"). If no number is specified, our bot will automatically subtitle the first video found.

Does the account need to be public?

Yes, SRTGenDotCom can only process tweets from public accounts. If the account or the specific tweet is private, our bot won't be able to fetch the content.

What if the reply is nested in a long thread?

To keep things accurate, SRTGenDotCom only looks for media in the immediate parent tweet. If the video you want to subtitle is further up the thread, please mention the bot in a direct reply to that specific post.

Why was my X quota used if the transcription failed?

We consume an 'Ask X' quota as soon as we make a request to the X API to process your mention, as this is a billed interaction on our end. However, your SRTGen transcription credits are protected—they are only deducted for successfully subtitled videos.

How does the credit system work?

Credits are the universal currency of SRTGen. We offer a transparent consumption model: 1 minute of AI Transcription costs 1 credit, while AI Translation costs 0.5 credits per minute, and Burn Video rendering costs only 0.25 credits per minute.

What happens if I run out of credits?

If you've used your monthly allotment, you can either purchase 'On-Demand' top-ups or upgrade to a higher plan with a larger credit allowance. When you upgrade, your old plan is automatically canceled, and your new credit balance is activated immediately.

Do my credits roll over to the next month?

Subscription credits are refreshed monthly and do not roll over. However, any 'On-Demand' credits (top-ups) you purchase are permanent and will remain in your account until you use them.

Can I cancel my subscription anytime?

Yes, you have full control. You can cancel your subscription at any moment through the billing portal. You will retain all your plan benefits and credits until the end of the current billing cycle.

What is your refund policy?

All purchases are generally non-refundable. Subscription and top-up credits are final once processed. However, we reserve the right to issue account credits or refunds at our sole discretion in exceptional cases, such as documented service outages or billing errors.