ScribVoiceWAV to Text
Convert WAV to text in minutes
Upload a WAV, WAVE or BWF recording, choose from 38 spoken languages, and review the transcript with speakers and timestamps when processing finishes. No conversion step is required before upload.
WAV carries no compression, so size comes first
WAV is an uncompressed PCM container. Its file size grows with the sample rate, bit depth, channel count and duration because the audio is stored without a lossy compression step.
Your plan sets the limit: anonymous trial 100 MB, Free 200 MB, Basic 1000 MB, Standard and Pro 2048 MB. The uploader shows the allowance for your plan, and paid tiers also have their own duration limits.
| Format | Approximate size per minute | What a 100 MB guest allowance holds |
|---|---|---|
| 44.1 kHz / 16-bit / stereo | about 10 MB | about 10 minutes |
| 48 kHz / 16-bit / stereo | about 11 MB | about 9 minutes |
| 96 kHz / 24-bit / stereo | about 33 MB | about 3 minutes |
| 8 kHz / 16-bit / mono (phone audio) | about 1 MB | the full 30-minute limit |
One ceiling is not set by the plan at all: a .bwf file is limited to 200 MB on every plan. At 96 kHz / 24-bit stereo that is roughly 6 minutes, while the same audio saved as .wav can use the full Standard or Pro allowance.
A normal high-sample-rate WAV therefore fills the guest allowance long before the 30-minute duration limit: about 10 minutes at 44.1 kHz/16-bit stereo and about 3 minutes at 96 kHz/24-bit stereo.
Only the phone-quality 8 kHz mono case can use the full duration allowance, because a minute of that audio is about 1 MB.
Read the sample rate, bit depth and channel count before uploading
- Sample rate: 8 kHz is typical of phone systems, 16 kHz is common for meetings and interview devices, 44.1 or 48 kHz is standard recording practice, and 96 kHz or higher usually comes from professional equipment or post-production work.
- Bit depth: 16-bit is most common, 24-bit appears in field recordings and engineering files, and 32-bit floating point is often an audio-software export.
- Channels: Mono files are smaller, but overlapping speakers can become harder to separate. Stereo is not automatically better; the original microphone arrangement matters.
- Extension: .wav and .wave are the same general container. .bwf is a WAV variant that carries broadcast metadata, but browsers cannot inspect this format, so .bwf is capped at 200 MB on every plan, including Standard and Pro. Check the home page's supported list before using unusual extensions such as .w64 or .rf64.
ScribVoice does not convert the file for you, and you do not need to standardize its sample rate or bit depth first. Limits are measured in bytes, so higher specifications reach the limit sooner.
Where WAV files actually come from
Field recorders and live interviews
These commonly use 44.1 or 48 kHz. Long high-rate recordings can become very large, but the transcript is useful for interview quotes, research notes and searchable field records.
Call centers and telephone systems
Phone audio is often 8 kHz mono and may use μ-law or A-law rather than PCM. Names, numbers and order codes are easy to misrecognize, so compare each important detail with the recording after transcription.
Audio software project exports
Audacity, Logic and Pro Tools can export 24-bit, 32-bit float, 96 kHz or multichannel WAV files. They make a high-quality source for editing and verification, but often create the largest files.
Meeting, legal and medical recordings
Organizations may keep an uncompressed original for traceability. A transcript makes those records easier to search, review and archive while preserving the source.
Podcast, radio and video production tracks
Editors often keep the original interview track as WAV before delivering an MP3 or AAC version. Transcription can support a script, quote search or rough cut before the final edit.
Supported upload formats are listed in the Supported uploads section of the home page.
8 kHz phone audio and 96 kHz masters are not the same problem
An 8 kHz mono call recording has a frequency ceiling of 4 kHz, so speech detail is limited and names, numbers, amounts and proper nouns are the most vulnerable. Transcribe it first, then check every critical field against the audio.
A 96 kHz / 24-bit master is not usually hard because of sound quality; it is hard because of size. Calculate the duration and estimated bytes, then choose which version to upload rather than changing its sound blindly.
A 48 kHz / 16-bit stereo recording is a practical middle case: adequate sound quality, manageable size and generally stable speaker separation for conversations.
The lower bound of transcription quality comes from the original frequency range and signal-to-noise ratio, not the sample-rate number alone. A clean 8 kHz call can be easier to recognize than a reverberant 96 kHz recording.
When the file is too large: trim, split, or upload a smaller copy
- 1. Trim first. Use free audio software to keep only the needed section. This costs no sound quality and also shortens the duration.
- 2. Split the recording. Cut a long session into chapters and transcribe each file, then join the results by their timeline. The cost is a little more file management.
- 3. Make a compressed copy. Only do this when the original is too large. An MP3 or M4A copy usually remains intelligible for speech, but retain the WAV as the reference source.
Do not repeatedly transcode or lower the sample rate to "compress" a WAV. WAV is uncompressed, so lowering the sample rate loses information without providing much size advantage.