ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS

Lizhi Wan; Shijun Luo

REVIEW 3 major objections 4 minor 18 references

ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read ASR roundtrip checks can certify TTS audio as correct even when human listeners hear a wrong reading.

desk verdict Useful, well-scoped demonstration that ASR-roundtrip can mask human-perceived reading errors in Chinese news TTS; treat the 46/110 as a yield, not a rate. read the letter →

arxiv 2608.10606 v1 pith:EY3UJBWY submitted 2026-08-11 cs.CL

Shijun Luo , Lizhi Wan This is my paper · ORCID

classification cs.CL

keywords text-to-speechevaluationASRroundtripChinesenewsTTSreadingerrorscontext-dependentdecisionstextnormalizationspeechpolyphonedisambiguation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper targets a blind spot in a cheap, widely used text-to-speech check: synthesize speech, run it through speech recognition, and compare the transcript to the source text. It argues that for Chinese news spans whose correct reading depends on context or domain conventions, such as sports scores, aircraft models, technical units, and membership names, this ASR-roundtrip can return the intended written form even when the spoken audio is wrong. A human audio-first audit of 110 high-risk cases confirms 46 such masked errors for one TTS system and 51 for a second, reported with the full denominator of exposed-error and no-error cases. The paper concludes that roundtrip transcription remains useful for screening but should not stand alone as ground truth for reading-risk evaluation.

What carries the argument

The load-bearing object is the Context-Dependent Reading Decision (CDRD) span: a written string whose correct spoken form depends on context beyond the local characters, such as whether a hyphen means a score, a range, a model number, or something else. The mechanism that masks errors is ASR surface recovery, in which the recognizer's language-model and text-normalization priors map fluent-but-wrong audio back to the conventional written form. The paper also uses a span-isolation diagnostic that cuts a target span out of its sentence, showing that full-sentence context is what pushes many transcripts back to surface-correct.

What would settle it

Run the same audio-first span audit on a random sample of Chinese news TTS clips that is not pre-filtered for high-risk forms and compare the masked-error rate to the transcript-implied error rate; if ASR roundtrip never produces a surface-correct transcript for a human-confirmed wrong reading in that sample, the paper's central claim would be refuted. Alternatively, if an ASR decoder with all normalization and language-model recovery disabled still recovers the surface form in the same 46 cases, the effect would be an artifact of the protocol rather than of ASR priors.

Watch

Extended reading notes

Core claim

The central discovery is that ASR-based roundtrip evaluation produces false negatives precisely for fluent reading errors that listeners do notice. When a TTS frontend misreads 13-11 as a range instead of a snooker score, or 伊尔-76 as a negative number instead of an aircraft model, the ASR decoder's contextual and normalization priors can reconstruct the intended or surface-correct text, hiding the error. Evidence comes from a 110-case targeted audit in which the primary annotator heard wrong Raw TTS in 55 of the MiMo cases (46 masked plus 9 exposed), with the remaining 55 clean; a CosyVoice run on the same pool found 51 masked cases and 27 exposed errors. Isolating the span from sentence context re-exposed 18 of the 46 MiMo masked cases, and across the 97 confirmed-masked files Qwen3-ASR recovered surface-correct text in 40 while Paraformer did so in only 2, showing that masking is ASR- and protocol-dependent.

Load-bearing premise

The audit's category counts rest on one primary listener's judgment that the expected readings define correctness, for example that 13-11 must be read as a snooker score with 比, and that judgment was checked on only a 30-case subset with 0.634 agreement, so if those conventions are disputed the 46 of 110 numbers shift.

Editorial extensions

If this is right

For Chinese news TTS, transcript-only scores should be reported alongside human span-level listening whenever CDRD or CDRD-adjacent material is at stake.
Masking concentrates in scores, aircraft models, and unit strings, so evaluations built on news of those kinds need audio-first checks.
A span-isolation pass can serve as a cheap diagnostic that re-exposes many masked reading errors, recovering 18 of 46 in this audit.
Roundtrip comparison remains appropriate for screening and ablation but should be treated as an upper-bound sanity check, not standalone ground truth.
The choice of ASR system changes the measured masking rate, so cross-ASR controls are needed before drawing conclusions from roundtrip scores.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper's claims, a random-sample audit would be needed to estimate a production masked-error rate; the 46 of 110 figure comes from a deliberately high-risk pool and should not be quoted as prevalence.
We would expect the same surface-recovery route to appear in any modern Chinese news TTS pipeline that feeds short numeric or mixed-script spans without explicit spoken-form rewriting, although the paper only documents two TTS systems.
The large gap between Qwen3-ASR and Paraformer recovery rates suggests that newer decoder-based ASR systems may be more prone to normalizing away reading errors, an implication the paper does not explicitly draw.
A natural testable extension is to apply the span-isolation diagnostic to convention-heavy spans in other languages, such as sports scores and unit strings in English or Japanese, to see whether this masking is a general speech-evaluation phenomenon.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read this one. It makes a narrow claim and mostly supports it: ASR-roundtrip can miss reading errors a human listener would catch in Chinese news TTS, in spans where the correct reading depends on context or convention (scores, aircraft models, units, membership names). The evidence that the failure mode exists is solid—46 confirmed masked cases out of a 110-case targeted audit with human-audited wrong audio and surface-correct ASR transcript, plus 9 exposed errors and 55 no-error cases, so the denominator is complete and the labeling is not positive-only. Cross-TTS (CosyVoice) and cross-ASR (Whisper, Paraformer, Qwen3-ASR) controls add real weight. They also release the benchmark, transcripts, audio, and scoring code, so the result is checkable. The CDRD taxonomy and the frozen 200-case benchmark are new and useful for the speech evaluation subfield.

I agree with the stress-test note, with one clarification: the flaw is not in the existence claim, it is in the headline count. The 46/110 is a yield from a targeted pool, not a prevalence or error rate. The pool was mined using suspected-masking signals, including ASR transcripts that already look surface-correct, so 46 is an upper-bound-ish demonstration of mechanism, not a measured frequency. The ground truth is one primary annotator applying predefined expected readings (e.g., 13-11 must be 十三比十一). Validation on 30 cases gives exact-label κ=0.634 and binary κ=0.800; that is real but modest, and 80 of the 110 rows have no independent label. So the specific counts should be cited cautiously, and the paper mostly says this itself—it repeatedly labels the numbers as targeted audit yields and explicitly says the experiments establish mechanism, not production incidence. Good.

Where I push back on the stress-test note a little: calling it 'not robustly established' might be too strong for the existence claim. Even one verified wrong-audio/surface-ASR case establishes the failure mode; 46 such cases with a complete denominator and a documented protocol is more than enough. I would frame it as 'the existence claim is established; the rate is not.'

The bigger concern is the span-isolation diagnostic crossing ASR routes (full-context uses the original audit route, clips use strict ASR). The paper discloses this. It weakens the isolation result as a clean mechanism argument, but the Qwen in-house control partially addresses it.

Bottom line: this is a solid, honest paper for the speech evaluation community. It deserves peer review. I would accept it with a request to tone down rate language and add a second annotator on the full 110 or at least a larger random subset. A serious referee should engage.

Referee Report

3 major / 4 minor

Summary. The paper documents a failure mode of ASR-roundtrip evaluation for Chinese news TTS: short written spans whose correct reading depends on context or domain conventions (sports scores, aircraft models, units, membership names) can be read incorrectly by a TTS system while ASR transcribes the audio as the intended or surface-correct text. The authors construct a 200-case frozen benchmark from production news scripts and synthetic cases, define three risk labels (CDRD-entity, CDRD-polyphone, CDRD-adjacent), run Raw and oracle-style Structured text conditions with MiMo and CosyVoice TTS, and perform human span audits. A targeted 110-case audit reports 46 confirmed masked cases, 9 exposed TTS errors, and 55 no-error cases for MiMo; context isolation re-exposes 18/46; a Raw-only CosyVoice audit finds 51 masked cases; and cross-ASR comparisons show strong evaluator dependence. The central claim is that ASR-roundtrip can miss listener-perceived reading errors and should not be treated as standalone ground truth for Chinese news reading-risk evaluation.

Significance. If the existence claim holds, this is a useful and falsifiable caution for the TTS evaluation community. The paper has several strengths: the targeted audit is reported with a complete denominator rather than positive-only examples; the main failure cases are documented with human listening judgments and concrete transcripts; the authors include cross-TTS and cross-ASR controls; and they release prompts, settings, transcripts, labels, audio, summaries, and scoring code on GitHub and Zenodo. The paper also explicitly distinguishes targeted-audit yields from production prevalence and labels the Structured condition as an oracle-style diagnostic rather than a deployable frontend. These design choices make the central mechanism credible even though some headline counts need more careful qualification.

major comments (3)

[§3.3, Table 1] The headline count of 46 confirmed masked cases in Table 1 rests on the primary annotator's audio judgments for the full 110-case pool, but the blind relabel subset consists of only 30 rows and achieves 23/30 exact full-label agreement (κ=0.634) and 27/30 binary masked-versus-other agreement (κ=0.800). For the exact four-category partition that defines the 46/9/55 counts, this is moderate agreement, and the remaining 80 rows have no independent labels. The existence claim survives—one verified case would establish the mechanism—but the magnitude '46 confirmed masked' should be presented as protocol-dependent, ideally with per-annotator counts or a sensitivity analysis, rather than as an unqualified confirmation.
[§3.3, §4.1, Abstract] The 110-case pool is selected using suspected masking signals that include ASR transcripts that already appear surface-correct, so the counts in Table 1 are yields conditional on a non-random screening procedure, not rates or prevalence estimates. The paper does state this in §4.1 and §5, but the abstract's 'complete denominator' framing and Table 1's 'total targeted pool' column can easily be misread as incidence. I recommend removing 'complete denominator' from the abstract or explicitly writing 'among 110 pre-selected high-risk candidates' whenever the 46/110 count is invoked.
[§4.3, Table 3] The claim that context isolation re-exposes 18/46 masked errors is not a clean context manipulation for MiMo: the original full-sentence condition uses the case-specific audit route, while the clips use the main strict ASR route, so the ASR protocol changes together with the context. The same-decoder Qwen control in §4.5 provides supporting evidence for the isolation mechanism, but Table 3 should be relabeled as a combined context-plus-route comparison, or a same-route isolation comparison should be reported as the primary evidence.

minor comments (4)

[§3.2] The text contains a typo in 'oﬀicial score' that should read 'official score'.
[Abstract and §1] The phrase 'insuﬀicient' should be corrected to 'insufficient'.
[§3.3] The description of the blind relabel subset would be clearer if it stated whether the blind annotators saw the same predefined expected readings and negative readings as the primary annotator, since this affects how the reported κ values should be interpreted.
[§4.5, Table 4] The table would benefit from a note that the 0/46 and 2/51 counts are occurrence-aware recovery counts on previously confirmed masked files, not full-corpus error rates, to prevent readers from interpreting them as ASR quality rankings.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the masking claim rests on external linguistic conventions, actual TTS/ASR outputs, and disclosed human audio judgment, not on the paper's own definitions.

full rationale

The paper's central claim is that ASR-roundtrip can return a surface-correct or expected transcript while the TTS audio realizes a wrong reading. This is established empirically: the authors run real TTS systems, transcribe with real ASR systems, and have human annotators judge the audio against predefined expected readings such as 13-11 being read as 十三比十一. The expected readings are external linguistic and domain conventions, not quantities fitted or derived from the TTS or ASR outputs, so the masked-error categories are not self-definitional. The 'confirmed masked' label requires both a human audio judgment that Raw TTS is wrong and at least one ASR route transcribing the expected/surface-correct form; both conditions are independently observable outcomes. The Structured condition is explicitly described as an oracle-style diagnostic and is not used for the headline 46-case count; the paper discloses that the predefined readings define both the benchmark targets and the Structured condition. The targeted 110-case pool is nonrandom and selected using suspected masking signals, but this is a sampling limitation affecting whether 46/110 is a prevalence estimate, not a circular derivation. The IAA results (κ = 0.634 on 30 cases) are a reliability concern for the exact counts, but they do not make the existence claim circular. No load-bearing self-citation, imported uniqueness theorem, or ansatz-smuggling citation is present. The paper explicitly frames its results as mechanism evidence rather than production incidence, further reducing any risk that a fitted or definitional quantity is being passed off as an independent prediction.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim relies on domain conventions for correct readings and on human annotation as ground truth; these are reasonable but not externally validated. The targeted selection is disclosed and does not affect the existence claim.

assumptions (4)

domain assumption The paper's predefined expected readings for CDRD spans constitute the correct readings listeners expect.
The audit labels 'wrong' and 'correct' TTS readings based on conventions in Section 3.1 and the benchmark definitions in Section 2; these linguistic assumptions are not independently validated by an external standard.
domain assumption Human listening judgments by the primary annotator are a reliable proxy for 'listeners'.
The 46/110 count rests on one primary annotator with a 30-case IAA subset (Section 3.2); no large-scale listener study is performed.
ad hoc to paper The 110-case targeted pool selection using suspected masking signals does not invalidate the existence claim.
Section 3.3 selects candidates with suspected masking signals, so the count is not a prevalence estimate; the paper states it demonstrates mechanism, not prevalence (Discussion).
domain assumption ASR transcript normalization (removing whitespace, hyphens, colons, middle dots) is sufficient to judge surface-correct recovery.
Section 4.1 defines surface-correct recovery via this normalization; other normalizations might change the counts.

how reviews work

0 comments

Cite this review

Pith. "Pith review of ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS." pith.science (2026). https://pith.science/paper/EY3UJBWY

@misc{pith2026260810606,
  author       = {Pith},
  title        = {Pith review of: ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EY3UJBWY}},
  note         = {Machine review of arXiv:2608.10606}
}

read the original abstract

ASR-roundtrip evaluation is widely used as a scalable proxy for text-to-speech (TTS) intelligibility, but it can produce false negatives for reading errors perceived by listeners. We study Chinese news TTS spans whose correct reading depends on context or domain conventions, such as sports scores, aircraft models, technical units, and membership names. In these cases, Raw TTS can choose a plausible but wrong reading while ASR transcribes the audio as the intended or surface-correct text. A targeted audit over 110 high-risk MiMo TTS cases, reported with a complete denominator, confirms 46 masked false negatives, 9 exposed TTS errors, and 55 cases with no Raw TTS error. A span-isolation diagnostic re-exposes 18/46 previously masked errors. A Raw-only CosyVoice audit on the same targeted pool confirms 51 masked cases. Across the 97 TTS-specific audio files labeled confirmed masked across the two audits, Qwen3-ASR surface-recovers 40 cases, whereas Paraformer does so in only 2. The results suggest that ASR-roundtrip is useful for screening but insufficient as standalone ground truth for Chinese news reading-risk evaluation.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 16 canonical work pages

[1]

It scales cheaply and often tracks intelligibility, but its reliability depends on the task and protocol [1, 2]

INTRODUCTION Automatic speech recognition (ASR) roundtrip evaluation is attractive for TTS: synthesize speech, transcribe it, and com- pare the transcript to a reference. It scales cheaply and often tracks intelligibility, but its reliability depends on the task and protocol [1, 2]. This paper shows a specific failure mode for Chinese news TTS: the audio ...

work page
[2]

Index Terms— text-to-speech evaluation, ASR-roundtrip, Chinese news TTS, reading errors, speech evaluation

The results suggest that ASR-roundtrip is useful for screen- ing but insuﬀicient as standalone ground truth for Chinese news reading-risk evaluation. Index Terms— text-to-speech evaluation, ASR-roundtrip, Chinese news TTS, reading errors, speech evaluation

work page
[3]

ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS

READING-RISK BENCHMARK 2.1. Risk definition A CDRD span is a written span x whose correct reading de- pends on context c. The same local pattern can require dif- ferent readings across domains: a hyphen may mark a score, range, model, or subtraction-like form; a mixed digit-letter span may be a unit, product name, or membership tier. CDRD overlaps with Ch...

work page Pith review arXiv 2026
[4]

EV ALUATION PROTOCOL 3.1. Systems The primary system pairs MiMo-V2.5-TTS API synthesis with the audio-capable MiMo mimo-v2.5 API [7] prompted for verbatim transcription; mimo-v2-omni is used as a fall- back and protocol-ablation route. The strict ASR protocol mainly targets spoken numeral preservation and discourages transcript normalization; it reduces b...

work page
[5]

ASR false negatives in targeted audits Table 1 reports the targeted audit, with a complete denomina- tor, that supports the ASR false-negative claim

RESULTS 4.1. ASR false negatives in targeted audits Table 1 reports the targeted audit, with a complete denomina- tor, that supports the ASR false-negative claim. In the MiMo audit, human listening confirms 46 cases where Raw TTS is wrong and ASR transcribes the audio as the expected or surface-correct text. The same audit explicitly accounts for 9 expose...

work page
[6]

This does not imply that ASR always masks such errors, that 46/110 is a natural production rate, or that Structured is a deployable frontend

DISCUSSION The central claim is deliberately narrow: ASR-roundtrip can miss reading errors perceived by listeners in CDRD and CDRD-adjacent spans. This does not imply that ASR always masks such errors, that 46/110 is a natural production rate, or that Structured is a deployable frontend. ASR-roundtrip remains useful for scalable screening and ablation, bu...

work page
[7]

Human annotators performed span-level listening audits of synthetic audio and authorized the use of their anonymized judgments for this study

COMPLIANCE WITH ETHICAL STANDARDS This study uses news text, synthesized speech, ASR tran- scripts, and human listening judgments for TTS evaluation. Human annotators performed span-level listening audits of synthetic audio and authorized the use of their anonymized judgments for this study. No sensitive personal data, medi- cal data, biometric identifica...

work page
[8]

Confidence intervals for ASR-based TTS evaluation,

J. Taylor and K. Richmond, “Confidence intervals for ASR-based TTS evaluation,” Proc. Interspeech, 2021

work page 2021

Show all 18 references

[9]

Automatic human utility evaluation of ASR systems: Does WER really predict performance?

B. Favre et al., “Automatic human utility evaluation of ASR systems: Does WER really predict performance?” Proc. Interspeech, 2013

2013
[10]

An end-to-end Chinese text normalization model based on rule-guided flat-lattice Transformer,

W. Dai et al., “An end-to-end Chinese text normalization model based on rule-guided flat-lattice Transformer,” Proc. ICASSP, 2022

2022
[11]

Label embedding for Chinese grapheme-to-phoneme conver- sion,

E. Choi, H.- Y . Kim, J.-H. Kim, and J.-M. Kim, “Label embedding for Chinese grapheme-to-phoneme conver- sion,” Proc. Interspeech, 2021

2021
[12]

Normalization of non-standard words,

R. Sproat et al., “Normalization of non-standard words,” Computer Speech & Language, 2001

2001
[13]

The Kestrel TTS text normal- ization system,

P . Ebden and R. Sproat, “The Kestrel TTS text normal- ization system,” Natural Language Engineering, 2015

2015
[14]

Xiaomi MiMo Audio Un- derstanding API Documentation,

Xiaomi MiMo, “Xiaomi MiMo Audio Un- derstanding API Documentation,” 2026. [Online]. Available: https://mimo. mi.com/docs/en-US/quick-start/ usage-guide/multimodal-understanding/ audio-understanding. Accessed: Jul. 17, 2026

2026
[15]

Robust speech recognition via large- scale weak supervision,

A. Radford et al., “Robust speech recognition via large- scale weak supervision,” Proc. ICML, 2023

2023
[16]

CosyVoice: A scalable multilingual zero- shot text-to-speech synthesizer based on supervised se- mantic tokens,

Z. Du et al., “CosyVoice: A scalable multilingual zero- shot text-to-speech synthesizer based on supervised se- mantic tokens,” arXiv:2407.05407, 2024

2024 arXiv
[17]

Paraformer: Fast and accurate parallel Transformer for non-autoregressive end-to-end speech recognition,

Z. Gao, S. Zhang, I. McLoughlin, and Z. Y an, “Paraformer: Fast and accurate parallel Transformer for non-autoregressive end-to-end speech recognition,” Proc. Interspeech, pp. 2063–2067, 2022

2022
[18]

Qwen3-ASR technical report,

X. Shi et al., “Qwen3-ASR technical report,” arXiv:2601.21337, 2026

2026 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

It scales cheaply and often tracks intelligibility, but its reliability depends on the task and protocol [1, 2]

INTRODUCTION Automatic speech recognition (ASR) roundtrip evaluation is attractive for TTS: synthesize speech, transcribe it, and com- pare the transcript to a reference. It scales cheaply and often tracks intelligibility, but its reliability depends on the task and protocol [1, 2]. This paper shows a specific failure mode for Chinese news TTS: the audio ...

work page

[2] [2]

Index Terms— text-to-speech evaluation, ASR-roundtrip, Chinese news TTS, reading errors, speech evaluation

The results suggest that ASR-roundtrip is useful for screen- ing but insuﬀicient as standalone ground truth for Chinese news reading-risk evaluation. Index Terms— text-to-speech evaluation, ASR-roundtrip, Chinese news TTS, reading errors, speech evaluation

work page

[3] [3]

ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS

READING-RISK BENCHMARK 2.1. Risk definition A CDRD span is a written span x whose correct reading de- pends on context c. The same local pattern can require dif- ferent readings across domains: a hyphen may mark a score, range, model, or subtraction-like form; a mixed digit-letter span may be a unit, product name, or membership tier. CDRD overlaps with Ch...

work page Pith review arXiv 2026

[4] [4]

EV ALUATION PROTOCOL 3.1. Systems The primary system pairs MiMo-V2.5-TTS API synthesis with the audio-capable MiMo mimo-v2.5 API [7] prompted for verbatim transcription; mimo-v2-omni is used as a fall- back and protocol-ablation route. The strict ASR protocol mainly targets spoken numeral preservation and discourages transcript normalization; it reduces b...

work page

[5] [5]

ASR false negatives in targeted audits Table 1 reports the targeted audit, with a complete denomina- tor, that supports the ASR false-negative claim

RESULTS 4.1. ASR false negatives in targeted audits Table 1 reports the targeted audit, with a complete denomina- tor, that supports the ASR false-negative claim. In the MiMo audit, human listening confirms 46 cases where Raw TTS is wrong and ASR transcribes the audio as the expected or surface-correct text. The same audit explicitly accounts for 9 expose...

work page

[6] [6]

This does not imply that ASR always masks such errors, that 46/110 is a natural production rate, or that Structured is a deployable frontend

DISCUSSION The central claim is deliberately narrow: ASR-roundtrip can miss reading errors perceived by listeners in CDRD and CDRD-adjacent spans. This does not imply that ASR always masks such errors, that 46/110 is a natural production rate, or that Structured is a deployable frontend. ASR-roundtrip remains useful for scalable screening and ablation, bu...

work page

[7] [7]

Human annotators performed span-level listening audits of synthetic audio and authorized the use of their anonymized judgments for this study

COMPLIANCE WITH ETHICAL STANDARDS This study uses news text, synthesized speech, ASR tran- scripts, and human listening judgments for TTS evaluation. Human annotators performed span-level listening audits of synthetic audio and authorized the use of their anonymized judgments for this study. No sensitive personal data, medi- cal data, biometric identifica...

work page

[8] [8]

Confidence intervals for ASR-based TTS evaluation,

J. Taylor and K. Richmond, “Confidence intervals for ASR-based TTS evaluation,” Proc. Interspeech, 2021

work page 2021